« Development Centre Studies

Electronic Commerce
for Development

INTERNATIONAL DEVELOPMENT

Edited by Andrea Goldstein and

David O’Connor
 Development Centre Seminars

Electronic Commerce
for Development

Edited by
Andrea Goldstein and David O’Connor

DEVELOPMENT CENTRE OF THE ORGANISATION

FOR ECONOMIC CO-OPERATION AND DEVELOPMENT
 ORGANISATION FOR ECONOMIC CO-OPERATION
AND DEVELOPMENT

Pursuant to Article 1 of the Convention signed in Paris on 14th December 1960, and
which came into force on 30th September 1961, the Organisation for Economic Co-operation
and Development (OaECD) shall promote policies designed:
– to achieve the highest sustainable economic growth and employment and a rising
standard of living in Member countries, while maintaining financial stability, and thus
to contribute to the development of the world economy;
– to contribute to sound economic expansion in Member as well as non-member
countries in the process of economic development; and
– to contribute to the expansion of world trade on a multilateral, non-discriminatory
basis in accordance with international obligations.
The original Member countries of the OECD are Austria, Belgium, Canada, Denmark,
France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, the Netherlands, Norway,
Portugal, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States.
The following countries became Members subsequently through accession at the dates
indicated hereafter: Japan (28th April 1964), Finland (28th January 1969), Australia
(7th June 1971), New Zealand (29th May 1973), Mexico (18th May 1994), the Czech Republic
(21st December 1995), Hungary (7th May 1996), Poland (22nd November 1996), Korea
(12th December 1996) and the Slovak Republic (14th December 2000). The Commission of
the European Communities takes part in the work of the OECD (Article 13 of the OECD
Convention).
The Development Centre of the Organisation for Economic Co-operation and Development was
established by decision of the OECD Council on 23rd October 1962 and comprises twenty-two Member
countries of the OECD: Austria, Belgium, Canada, the Czech Republic, Denmark, Finland, France, Germany,
Greece, Iceland, Ireland, Italy, Korea, Luxembourg, Mexico, the Netherlands, Norway, Portugal, Slovak
Republic, Spain, Sweden, Switzerland, as well as Argentina and Brazil from March 1994, Chile since
November 1998 and India since February 2001. The Commission of the European Communities also takes
part in the Centre’s Advisory Board.
The purpose of the Centre is to bring together the knowledge and experience available in Member
countries of both economic development and the formulation and execution of general economic policies; to adapt
such knowledge and experience to the actual needs of countries or regions in the process of development and to
put the results at the disposal of the countries by appropriate means.

THE OPINIONS EXPRESSED AND ARGUMENTS EMPLOYED IN THIS PUBLICATION ARE THE SOLE
RESPONSIBILITY OF THE AUTHORS AND DO NOT NECESSARILY REFLECT THOSE OF THE OECD, THE
DEVELOPMENT CENTRE OR THE GOVERNMENTS OF THEIR MEMBER COUNTRIES.

*
* *
Publié en français sous le titre :
Commerce électronique et développement

© OECD 2002
Permission to reproduce a portion of this work for non-commercial purposes or classroom use should be
obtained through the Centre français d’exploitation du droit de copie (CFC), 20, rue des Grands-Augustins,
75006 Paris, France, tel. (33-1) 44 07 47 70, fax (33-1) 46 34 67 19, for every country except the United States.
In the United States permission should be obtained through the Copyright Clearance Center, Customer
Service, (508)750-8400, 222 Rosewood Drive, Danvers, MA 01923 USA, or CCC Online: www.copyright.com. All
other applications for permission to reproduce or translate all or part of this book should be made to
OECD Publications, 2, rue André-Pascal, 75775 Paris Cedex 16, France.
 Foreword

This publication falls within the 2001-2002 Development Centre work

programme under the title “Globalising Technologies and Domestic Entrepreneurship
in Developing Countries”. The project has produced a number of outputs as well as
this book whose chapters are derived from papers delivered to a May 2001 conference
organised in co-operation with the School of Development and Institutional Change
of the University of Bologna, under the patronage of Italy’s Ministry for Foreign
Affairs.

3
 Acknowledgements

The Development Centre and the School for Development and Institutional
Change (SDIC) acknowledge the support of Telecom Italia Lab, Fondazione Cassa di
Risparmio in Bologna, Endaveours and Sun in organising the conference held in
Bologna on 4-5 May 2001 under the patronage of the Italian Ministry for Foreign
Affairs. The interest and support received from the Government of Italy in the context
of this project is gratefully acknowledged.

4
 Table of Contents

Acknowledgements ......................................................................................................... 4
Preface ....................................................................................................................... 7
An Introduction to the Debate on Electronic Commerce and Development
Andrea Goldstein and David O’Connor ................................................... 9

PART ONE
INTERNET AND THE ORGANISATION
OF GLOBAL MARKETS

Chapter 1 The Evolution of Global Value Chains in the Internet Era

Gary Gereffi ................................................................................................. 1 9

Chapter 2 E-Commerce for Development: A General Framework

Patrizia Fariselli .......................................................................................... 3 5

Chapter 3 E-Commerce for Development: Between Scylla and Charybdis

David O’Connor .......................................................................................... 5 5

PART Two
SECTORAL ANALYSES

Chapter 4 The Prospects and Challenges of E-Business for the South African
Automotive Components Sector: Preliminary Findings from Two
Benchmarking Clubs
Sagren Moodley ........................................................................................... 6 7

5
Chapter 5 Local Entrepreneurship in the Era of E-business: Early Evidence
from the Indian Automobile Industry
Andrea Goldstein ......................................................................................... 9 3

Chapter 6 The World’s First Internet Coffee Auction: Design, Implementation

and Lessons Learned
Morten Scholer ............................................................................................ 121

PART THREE
COUNTRY EXPERIENCES AND LOCAL EXPERIMENTS

Chapter 7 The Micro-Foundations of E-Commerce: Informational-Focused

Development in Andhra Pradesh, India
Kyle Eischen ................................................................................................. 129

Chapter 8 Cellular Phones in Rural Bangladesh: A Study of the Village Pay

Phone of Grameen Bank
Salahuddin M. Aminuzzaman ..................................................................... 161

Chapter 9 Local Content Creation and E-Commerce: A South African Perspective

Carey-Ann Jackson and Johan Eksteen .................................................... 179

List of Participants ........................................................................................................... 201

Conference Agenda .......................................................................................................... 202

6
 Preface

In the last part of the past century, the application of information and
communication technologies (ICT), including the internet, has become widespread in
global business. This has been made possible by technical progress as well as policy
changes such as market opening to trade and foreign investment and service market
liberalisation, notably in telecommunications. Although new technologies appear to
be improving economic performance and welfare among user populations, in the
OECD countries the link between ICTs and economy-wide productivity advance has
been notoriously elusive. Early studies for developing countries also fail to detect a
significant association between economic growth and use of ICTs. A possible hypothesis
is that investing in these technologies bear fruit only once certain networks and
organisational changes are implemented. Individual firms investing in ICT, particularly
internet-based management systems, at this early stage in the development of e-business
might find the return on investment quite low. Over time, as more suppliers and
customers use the internet in their front-office and back-office systems and to connect
with their trading partners, the benefits of internet-enabled collaboration will likely
become more pronounced.
What does then the widespread diffusion of the internet and web-based
e-commerce imply for developing countries? Can their small producers use these
technologies to break into high-value segments and increase profitability and
competitiveness? Or, on the contrary, do they increase “commoditisation” and
strengthen the bargaining power of global buyers, mostly based in the OECD countries?
These questions are of increasing relevance to policy makers everywhere. Poorer
countries must devise appropriate strategies to turn new technologies into building
blocks towards their insertion into the global economy, but their OECD partners must
make sure that erecting barriers in the virtual marketplace does not offset the support
provided through development assistance.
The linkages between the internet and global market structures are far from
clear. The OECD’s pioneering role in developing e-commerce policy led the
Development Centre in 2000 to start research on the development implications of the
application of internet in business practices. This project, included in the 2001-02
programme of work under the title “Globalising Technologies and Domestic
Entrepreneurship in Developing Countries” has so far produced a number of outputs.

7
Two Technical Papers have been presented by Development Centre staff in Brazil,
Chile, Italy, Japan, Lebanon, Malaysia, Mexico and Poland. A joint op-ed piece on
the Financial Times and other still unpublished papers have also been produced. In
May 2001 a conference was organised in co-operation with the School of Development
and Institutional Change at, as I remarked at the time, the world’s oldest university in
Bologna under the patronage of Italy’s Ministry for Foreign Affairs. On this occasion,
scholars, practitioners and policy makers from both OECD Member and non-Member
countries analysed different questions associated to e-commerce, trying to go beyond
the veil of the digital divide to better understand what can be expected from the new
medium in terms of development.
As multinational corporations integrate the internet into their cross-border
business operations, firms from developing countries run the risk of exclusion from
global value chains if they cannot establish electronic ties with their major business
partners. In this context, it would be useful to focus on the ways in which ICTs affect
the growth prospects of firms in specific sectors. Further research should concentrate
on how net-based applications change organisation routines, modify the pool of
corporate resources and interact with strategy in firms of different size, nationality,
ownership and industries. Where does the quality of ICT assets, including internet
access, make the greatest difference to firm and industry competitiveness?
Policy makers need such detailed analyses for the design of policy and investment
measures to upgrade ICT infrastructure and promote ICT diffusion. This volume is a
major contribution to an area of economic research which is still in its infancy, despite
the current preoccupation with ICT.

Jorge Braga de Macedo

President
OECD Development Centre
2 October 2002

8
 An Introduction to the Debate
on Electronic Commerce and Development
Andrea Goldstein and David O’Connor*

The “digital divide” literature focuses on low rates of internet penetration and
high costs of access in much of the developing world. Most developing countries do
still have weak information and communications technology (ICT) infrastructures as
well as limited systems know-how. Still, consistent with Moore’s law, ICT products
and services have experienced significant cost reductions over time, which is permitting
developing countries, including some of the poorest, to improve their access to and
use of modern ICT at an impressive rate. While low incomes do continue to pose a
constraint to broader internet diffusion, perhaps more crucial is the establishment of a
conducive competitive and regulatory environment in the telecommunications and
internet service provider (ISP) sectors.
Assuming that ICTs continue to become more affordable, what does this imply
for entrepreneurs in developing countries? How can they use these technologies to
their advantage? What does the widespread diffusion of the internet and web-based e-
commerce mean for their profitability and competitiveness? The linkages between the
internet and global market structures are far from clear. Do they make it easier for
small producers to break into high-value segments and compete on a global scale, or,
on the contrary, do they increase “commoditisation” and strengthen the bargaining
power of global buyers, mostly based in the OECD countries?
The novelty of the internet means that there is little historical evidence on which
to venture projections of future trends, even in the OECD countries. E-commerce is
evolving rapidly, with the corporate landscape continuously transformed by start-ups,
___________________________________
* Our thinking on these matters owes a lot to the comments received on presentations
of our paper “E-Commerce for Development: Prospects and Policy Issues” in Beirut
(UN Economic Commission for Western Asia), Brasília (Presidencia da República),
Cagliari (Italian Economic Association), Santiago (Chilean Academy of Diplomacy),
Singapore (Asian Development Bank Institute), Tokyo (Japanese Bank for
International Co-operation), Tijuana (Colegio de la Frontera Norte) and Warsaw
(Leon Kozminski Academy of Entrepreneurship and Management).

9
acquisitions and failures, and with new technologies coming to market almost daily.
Digital trade in goods and services already makes it easier for artisans, musicians and
other artists in developing countries to access world business-to-consumer (B2C)
markets, cutting out layers of middlemen and improving the creators’ bargaining
power. For some developing countries, B2C e-commerce may offer important export
as well as local market development opportunities. Globally, however, the biggest
potential appears to be in the business-to-business (B2B) market. It includes a broad
range of inter-company transactions such as wholesale trade, corporate purchases of
technology, parts, components and capital equipment, and trade in services, including
financial services such as insurance, commercial credit, bonds, securities and other
assets (Lucking-Reiley and Spulber, 2001). How can developing country entrepreneurs
position themselves to participate in B2B markets? The papers presented at the Bologna
Conference and published in this volume deal with both sorts of e-commerce, as well
as with the broader question of how the internet may be reconfiguring global value
chains and inter-firm relationships along them.

Internet and the Organisation of Global Markets

In theory at least, market exchange is impersonal, episodic and carried out at

arm’s length. All that matters is how much the seller is asking and how much the
buyer is offering. Incomplete information in the international market, however, creates
difficulty in matching buyers and sellers and interferes with the ability of prices to
allocate scarce resources. Information technologies have the potential to reduce such
market imperfections. Information-sharing networks among internationally dispersed
agents can improve the allocation of resources, although their effectiveness depends
on whether they include all potentially important market participants. If certain large
low-cost producers of a particular good do not have network access, potential benefits
may not reach consumers. While the internet is more inclusive than early-generation
proprietary electronic networks such as EDI (Electronic Data Interchange), it is still
far from universal. Moreover, the existence of a network technology infrastructure is
not itself a sufficient condition for the emergence of a durable social or business
network. That depends on repeated interactions through which parties build reputations
for trustworthiness and gain confidence in one another. This still poses a challenge for
prospective new entrants into established networks. Does physical disintermediation
— i.e. the shift from face-to-face to virtual transactions — render it easier or harder
for developing-country newcomers to enter global markets? This question lies at the
core of the debate on e-commerce and development, since to different answers
correspond contrasting predictions of the chances for poor countries to break through
the glass ceiling to prosperity.
In his contribution to this volume, Gary Gereffi compares the three main drivers
of economic globalisation in the latter half of the 20th century: investment by
transnational corporations, international trade and the internet. Producer-driven and

10
buyer-driven value chains characterise the phases of investment-based and trade-based
globalisation. The emergence of the internet holds the potential to undermine traditional
models of commodity-chain organisation, substituting one in which new “info-
mediaries” (e.g. owners of B2C or B2B portals) provide the main impetus to the
organisation of transactions. By making markets they erode the market power of
traditionally dominant commodity-chain participants (buyers or sellers). Gereffi also
presents two alternative scenarios. In one the internet becomes the ultimate instrument
of consumer sovereignty, putting enough information at consumers’ disposal to permit
arbitrage to squeeze sellers’ profits. In the other, the lead firms in existing value
chains, whether producer-driven or buyer-driven, largely capture the e-commerce
benefits. He suggests that, at present, this last scenario appears the most probable.
The experience of Italian small and medium scale enterprises (SMEs) was invoked
as potentially relevant to SMEs in developing countries. Annaflavia Bianchi, discussant
of Professor Gereffi’s paper, argued that such enterprises are unlikely to be excluded
entirely from the benefits of internet use. The technologies are broadly affordable and
becoming more so, and e-commerce application tools are available on a pay-for-use
basis, avoiding the obstacle of large fixed investment costs. The internet also makes
possible greater specialisation of functions and more division of labour in administrative
tasks (witness the outsourcing of back-office functions), which could perhaps benefit
small enterprises that lack expertise in, say, accounting, finance and marketing.
The literature on Italian industrial districts, a form of business network that has
attracted much interest in development circles, also offers possible lessons. The
competitive success of these districts is thought to lie in how firms become embedded
in supportive local socio-economic environments that allow them to overcome their
size limits in terms of managerial know-how, financial capital and technology. The
emergence of the internet raises the question whether such rich networks of supporting
institutions can be replicated in virtual space, or are too heavily dependent on
relationships of trust and the intensive personal interactions permitted by physical
(and cultural) proximity.
Patrizia Fariselli observes that the pervasive use of ICT infrastructures, services
and goods is changing the relationships between firms and the configuration of business
networks themselves. Established networks become unstable as interactivity increases,
global multipurpose technological platforms develop and business activities can be
de-composed and re-composed at low cost. Networking effects can give rise to increasing
returns, so that the standard models used to analyse productivity performance in the
“old economy” find their heuristic power diminished and the problems of measurement
become more difficult. Yet one recent attempt at measuring ICT-related network
economies (or productivity spillovers) for the US economy finds scant evidence of
their existence (Stiroh, 2001)1.
The internet’s most significant impact to date has been in B2B markets. E-
commerce is reshaping the competitive dynamics in traditional producer-driven and
buyer-driven value chains such as automobiles and coffee.

11
 Automobiles

The automotive industry has traditionally pioneered in new management and

organisational techniques, from mass production Taylorism in the 1920s to lean
manufacturing and just-in-time (JIT) in the 1980s. More recently automotive
manufacturers have reached a high level of sophistication in outsourcing the design,
manufacture and assembly of increasingly complex systems (e.g. Anti-blocking Brake
Systems, ABS) and modules (e.g. the vehicle cockpit), in the process narrowing the
immediate supply base to trusted Tier I suppliers. Car assemblers are also among the
largest multinational companies in terms of assets located outside their home markets,
and producers of motor vehicles are consistently ranked among the largest investors in
developing countries.
Two papers in this volume analyse the impact of internet technology in the car
industry in developing countries. Sagren Moodley examines the prospects and challenges
for the Kwa-Zulu Natal automotive components sector in South Africa. He notes that
reaching the ideal type of lean supply chain managed over the internet will be expensive
and requires a sort of industry-wide co-operation not generally prevailing in the South
African automotive components industry. The critical mass needed to reap positive
network externalities is not yet present in the sector. An important insight from this
study concerns the role of leadership by original equipment manufacturers (OEMs)
and the large Tier I suppliers in defining e-business standards for all value-chain
participants and in creating incentives that attract more companies to join digital
supply chains. Early adopters can also play a catalytic role in promoting the diffusion
of e-business in the components sector.
Research by Andrea Goldstein explores the use of e-business in the Indian car
industry, taking Fiat India as a case study and looking in particular at the effects on
domestic companies in the supply chain. Although net-based solutions are expected to
streamline the automotive supply chain, they have less importance than progress in
implementing lean manufacturing, which the internet can facilitate. Fiat produces a
homogeneous product for emerging markets in eight countries, including India, and
has been very successful in optimising supply-chain management in Brazil, where it
operates one of the largest non-OECD car factories. The auto business in India, and
supplier-OEM relationships in particular, are not immune from global pressures,
including oversupply, industry consolidation, modular assembly and the increasing
use of the internet. The last appears as a core element in the internal organisation of
major auto manufacturers such as Fiat and as taking a progressively more important
role in inter-firm relationships as well. Thus far, however, tangible effects on the
supply chain in India appear more in e-information management (knowledge
management, research and development and marketing) than in procurement.
Still, the auto industry has no unanimity on how the internet will shape the
organisation of supply chains and relations between OEMs and suppliers or between
different tiers of suppliers. To caricature, the industry has two contrasting supply-
chain models, one epitomised by Toyota and the other by General Motors (GM). The
Toyota model stresses durable supplier-customer relationships, in which OEM and

12
Tier 1 suppliers share major responsibilities for model design and development. The
GM model involves sourcing from the lowest-cost supplier, even if that means frequent
switching and repeated short-term contracts. These contrasting models are perhaps
more accurately used to describe, on the one hand, the relationship between OEMs
and Tier I suppliers (Toyota) and, on the other, the relationship between the first two
and lower tier suppliers (GM), where the latter make more standardised components
and parts. The internet may have the greatest impact in the second type of relationship
by fostering greater price competition among suppliers (e.g. through on-line bidding)
and pooling the demands of OEMs or other buyers to reap ever greater scale economies.
What remains unclear is whether the criteria for participation in on-line exchanges
such as Covisint — the exchange established by major auto OEMs and parts
manufacturers — will act as major barriers to entry to prospective new suppliers.
The discussant, John Humphrey, suggested that the threat posed by internet-
based procurement to longstanding customer relationships might be more perceived
than real, citing the example of the horticultural industry. There, UK consumers have
become more demanding of assurances that the fresh produce on their supermarket
shelves is produced according to certain standards (e.g. regarding use of pesticides,
GMOs, etc.). As a result, the wholesalers and retail chains are more inclined to stick
with trusted suppliers than to search for new, lower cost suppliers via the internet.

Coffee and other Commodities

Morten Scholer describes the first on-line coffee auction, highlighting the potential
that e-commerce holds for certain commodity producers in developing countries. The
first auction, for premium Brazilian coffees, was organised in 1999 with considerable
outside technical support, notably from the International Trade Centre (an UNCTAD/
WTO joint venture). The Brazilian industry saw value in repeating it in 2000 on a
purely commercial basis. The precise designs of auction mechanisms will vary by
commodity. The success of the Brazilian auction depended in part on sending samples
for tasting to pre-qualified bidders. Evaluation of physical samples may or may not
be critical for other commodities. Where grading is used to gauge the quality of an
auctionable commodity, the prospective bidders’ degree of trust in the graders is crucial.
The Brazilian coffees commanded above-average prices largely by virtue of their pre-
selection for quality. Conventional markets tend to undervalue such premium products.
For standardised commodities, the distribution of surplus may favour consumers rather
than producers, although that is not a foregone conclusion. An interesting question
raised by David Hallam, paper discussant, was what long-run effect the internet can
be expected to have on commodity price variability.

Country Experiences and Local Experiments

Several papers present experiences of ICT use in developing countries or their

administrative divisions, rather than in specific sectors of the economy. They
demonstrate the catalytic effect that governments and NGOs can have on ICT and

13
internet use. According to Kyle Eischen, the state government in Andhra Pradesh,
India, has pursued a pro-active ICT strategy, partly to build the infrastructure and
human capital base for attracting investments in the software and services industries
that, in earlier days, had gone mostly to Bangalore in Karnataka. Numerous government
services, including land titling and taxation, can now be transacted on-line, with
significantly reduced processing times. The move to e-government has generated a
local demand for network software developers.
The Grameen Phone experiment in Bangladesh is particularly well studied. Its
intent is to make village pay phones (VPP), based on cellular radio technology, available
in rural areas of Bangladesh. Salahuddin Aminuzzaman highlights as one interesting
feature of this experiment the reliance on the credit-evaluation capabilities and the
infrastructure of the Grameen Bank to facilitate selection of VPP operators and bill
collection. As with Grameen microcredit borrowers, the phone operators are
overwhelmingly women, while the users are still predominantly men. The VPPs appear
to have provided significant benefits, including reduced isolation and fragmentation
of village economies, extension of the boundaries of the local market, timely price
information and improved terms of trade for farmers and recipients of overseas
remittances, and faster communication between family members and enhanced kinship
bonding. The income generated by the phone business has improved the status of
women within the household, while their command of a wealth of valuable information
has increased their status outside the home. This “infomediary” function of the village
phone operator will likely take on added importance if the internet is to have much
value to the illiterate. A still under-explored aspect of the Grameen Bank/Grameen
Phone experience is the degree to which access to mobile telephony by poor rural
women can leverage investments made with micro-credits and/or open new investment
opportunities.
Carey-Ann Jackson and Johan Eksteen describe two government-supported pilot
projects in South Africa, emphasising that, in the particular context, creation of local
content should also be seen as part of a “nation-building” effort. One of the projects
supports the creation of micro-enterprises to add value to agricultural and other primary
commodities. One component (cited as an example of a “livelihoods approach” to e-
commerce) consists of developing an architecture for a virtual trade house for these
value-added products. Typical generic functions are designed into the virtual trade
house such as distribution, payments, optimisation, centralisation, auditing and customer
account management. The issue of branding must still be resolved, with implications
for the distribution of income from, for example, royalty payments to participating
communities. The project is still too new to enable impact evaluation. A second project
deals with cultural preservation, in part through the promotion of local content creation
for virtual cultural tourism. Jackson’s paper notes the tension between the need to
retain a degree of authenticity and making the web site as appealing as possible to the
virtual tourist. The latter becomes all the more important if the site is seen as an
important medium for promoting actual tourism.

14
The Road Ahead

The most powerful message from the Bologna conference says that as multinational
corporations integrate the internet into their cross-border business operations, firms
from developing countries run the risk of exclusion from global value chains if they
cannot establish electronic ties with their major business partners. Beyond this general
proposition, however, an evident need persists for detailed sectoral analyses of how
ICTs transform the way business is done and the place of developing country firms in
this “brave new world”.
Although new technologies appear to be improving economic performance and
welfare among user populations, in the OECD countries the link between ICTs and
economy-wide productivity advance has been notoriously elusive. Temple (2000),
for instance, observes that productivity gains have occurred mostly in sectors intensive
in ICT — e.g. the introduction of ATMs in the financial services sector — and have
consisted largely of gains in labour productivity rather than total factor productivity
(a point confirmed by Stiroh, 2001). An early study for developing countries also
fails to detect a significant association between economic growth and use of ICTs
(Rodríguez and Wilson, 2000). More recent work by Pohjola (2001) corroborates it.
A possible hypothesis is that the networks and organisational changes necessary to take
advantage of these technologies take time to implement (cf. Askenazy and Gianella, 2000).
Individual firms investing in ICT, particularly internet-based management systems, at
this early stage in the development of e-business might find the return on investment
quite low. Over time, as more suppliers and customers use the internet in their front-
office and back-office systems and to connect with their trading partners, the benefits
of internet-enabled collaboration will likely become more pronounced.
At this stage in the development of ICT in developing countries, a more fruitful
avenue of research than seeking to detect broad growth impacts would be to focus on
the ways in which ICTs affect the growth prospects of firms in specific sectors, given
a detailed knowledge of industry organisation and dynamics. Research should focus
on how net-based applications change organisation routines, modify the pool of
corporate resources and interact with strategy in firms of different size, nationality,
ownership and industries. Where does the quality of ICT assets, including internet
access, make the greatest difference to firm and industry competitiveness? Policy
makers need such detailed analyses when designing policy and investment measures to
upgrade ICT infrastructure and promote ICT diffusion.

Note

1. Stiroh finds that ICT capital investment is significantly related to labour productivity
but has no significant effect on total factor productivity. He argues that it if ICT
capital is different from other types of capital investment in terms of its productivity
effects, this should show up in TFP.

15
 Bibliography

A SKENAZY, P. and C. GIANELLA (2000), “Le paradoxe de productivité: les changements

organisationnels, facteur complémentaire à l’informatisation”, Économie et
Statistique, No. 339-340.
LUCKING-REILEY, D. and D. SPULBER (2001), “Business-to-Business Electronic Commerce”,
Journal of Economic Perspectives, Vol. 15, No. 1.
POHJOLA, M. (ed.) (2001), Information Technology, Productivity, and Economic Growth:
International Evidence and Implications for Economic Development, Oxford
University Press, Oxford.
R ODRÍGUEZ F and E.J. WILSON, III (2000), “Are Poor Countries Losing the Information
Revolution?”, INFODEV Working Paper, May, processed.
STIROH, K.J. (2002), “Are ICT Spillovers Driving the New Economy?”, The Review of
Income and Wealth, No. 1(48).
TEMPLE, J. (2000), “Summary of an Informal Workshop on the Causes of Economic Growth”,
OECD Economics Department, Working Papers No. 260.

16
 PART ONE

INTERNET AND THE ORGANISATION

OF GLOBAL MARKETS

17
 Chapter 1

The Evolution of Global Value Chains

in the Internet Era
Gary Gereffi*

Shifting Forms of Value-Chain Governance

There have been three main drivers of economic globalisation in the latter half
of the 20th century: investment by transnational corporations, international trade and
the internet. Each has expanded the scope of global integration by altering how people,
resources and places are connected in economic transactions. This can be seen in three
broad and to some degree overlapping phases of globalisation:
1) Investment-based globalisation (1950-70) — International production networks
were the primary vehicles for this form of globalisation. In the 1950s and 1960s,
the spread of transnational corporations accelerated in a growing number of
manufacturing and raw material industries, and the ability of global companies
to try to manage the world as an integrated unit was seen by some as a threat to
national sovereignty (Vernon, 1971; Barnet and Müller, 1974).
2) Trade-based globalisation (1970-95) — In the 1970s, there was a marked shift
to export-oriented industrialisation as a preferred development strategy in many
parts of the developing world, beginning with East Asia, but spreading in the
1980s to Latin America, Africa, and elsewhere (Gereffi and Wyman, 1990).
This shift in national development strategies towards exports was premised on
the rapid and diversified industrialisation of a wide range of developing nations.
3) Digital globalisation (1995 onward) — In the mid-1990s, the information
revolution and a growing acceptance of the internet began to create an explosion
in connectivity owing to the open and almost cost-free exchange of a widening
_______________________________
* This chapter originally appeared as an article in the IDS Bulletin, Vol. 32, No. 3, 2001.
Permission to reprint is gratefully acknowledged.

19
 universe of rich information (Evans and Wurster, 2000). Asymmetrical access
to the Internet has given rise to concerns about a “digital divide” between those
countries and firms that were connected to modern information technology, and
those that were not.
Globalisation has been defined as “not merely the geographical extension of
economic activity across national boundaries but also — and more importantly — the
functional integration of such internationally dispersed activities” (Dicken, 1998: 5,
italics in original). But what underlies this functional integration of globalisation
processes? At the most fundamental level, information and the mechanisms for
delivering it are the unifying force that holds together the structure of business. In the
investment-driven phase of globalisation, vertically integrated transnational corporations
relied on proprietary information systems and hierarchical control to extend their
global reach. In trade-based globalisation, the operations of transnational corporations
were decentralised and firms in different parts of the world allied with each other to
form global value chains that involved horizontal rather than vertical co-ordination,
and resembled spider webs more than pyramids (Reich, 1991: chapter 7). These value
chains are shaped by the same kind of informational logic found in vertically integrated
companies, but in a weaker form. The new digital era of globalisation is characterised
by a dramatic increase in connectivity that is melting the informational glue that holds
corporations and global value chains together.
Competition in the global economy is forged by the interaction of three broad
sets of factors: technological, institutional and organisational innovations; the enterprise
networks that emerge out of and increasingly move beyond national business systems;
and the regulatory powers vested in regional, national and local governments. These
patterns of competition change over time and across industries, and they are embodied
in the organisation of global value chains (Gereffi and Korzeniewicz, 1994; Gereffi
and Kaplinsky, 2001)1.
Governance has particular importance in global value chains (Humphrey and
Schmitz, 2001). It refers to the key actors in the chains who determine the inter-firm
division of labour and shape the capacities of participants to upgrade their activities.
Initially, only two types of governance structures predominated. Global value chains
were either producer-driven or buyer-driven (Gereffi, 1994). A third form, oriented
around the internet, began to emerge in the mid-1990s. The differences between these
value chains reflect major changes in how international production and trade systems
are organised over time, the emergence of new actors and economic roles in the
chains and a continuing shift in the locus of power from producers to retailers to
consumers. The concept of governance is neither static nor exclusive as a defining
feature of global industries.
The governance structures in global value chains must be understood in historical
perspective. Technological, institutional and organisational innovations, as well as
changes in regulatory environments, transform the structures of industries and the
power of the leading firms within them. Governance structures in global value chains

20
evolve in conjunction with the forces that shape industry structures. Thus, at a particular
time or within a given industry new governance structures co-exist and interact with
earlier forms. The second section of this article highlights the origins of producer-
driven and buyer-driven governance structures in global value chains, and the third
focuses on the current emergence of internet-oriented value chains. The paper concludes
by outlining several scenarios that explore the degree to which the internet challenges
or sustains existing governance structures.

An Evolutionary Perspective on Global Value Chains

Two fundamental changes in the international arena profoundly shape our

contemporary perspectives on global value chains. First, a widespread shift has occurred
in national development strategies from import-substituting industrialisation (ISI) to
export-oriented industrialisation (EOI) throughout the developing world (Gereffi and
Wyman, 1990). Buttressed by the policy prescriptions of powerful international
economic organisations such as the World Bank and the International Monetary Fund,
as well as the US government, this preference for EOI rests heavily on the experience
of the East Asian “miracle economies” from the 1960s to the mid-1990s. During this
period, Japan and a handful of other high-performing Asian economies (most notably,
the “four tigers” of Hong Kong, Chinese Taipei, South Korea and Singapore) attained
booming exports and lofty per capita growth rates. They did so against the backdrop
of relatively low income inequality, high educational attainment and record levels of
domestic saving and investment (World Bank, 1993). EOI remains the development
orthodoxy in much of the world, despite the financial crisis that wracked Asia in the
late 1990s and increasingly strident criticisms among both academics and political
activists (Gore, 2000) of the Washington Consensus that favoured EOI.
Second, a major transformation in the organisation of global production also
occurred in the latter half of the 20th century. Under ISI development strategies,
transnational corporations were the dominant economic actors. They were vertically
integrated and had a global reach through the operations of wholly owned subsidiaries
that extracted natural resources for export or engaged in local production for sale in
domestic markets around the world. Exchange between core and peripheral areas has
since become much more complex. The surge of imports in developed countries since
the 1970s reveals that the centre of gravity for the production and export of many
manufactures encompasses a very diverse array of newly industrialising nations in the
developing world. As the relatively advanced East Asian and Latin American economies
have moved toward more technology-intensive and skill-intensive exports, it has
become clear that “cheap labour” alone no longer adequately explains Third World
industrialisation. The recent expansion of global production and trade networks is
propelled by both producer-driven and buyer-driven value chains.

21
 There is an affinity between the transitions from ISI to EOI strategies and the
shift from producer-driven to buyer-driven global value chains. The ISI strategy,
which prevailed in Latin America for nearly five decades until the 1970s, was based
on producer-driven value chains. Transnational corporations, which have actively
tapped Latin America’s oil, mineral and agricultural resources since the 19th century,
were invited to establish more advanced manufacturing industries in the region,
beginning with automobile assembly plants in large countries such as Mexico, Brazil
and Argentina in the 1920s. By the 1950s and 1960s, a range of advanced ISI factories
had spread throughout the region in diverse industries such as petrochemicals,
pharmaceuticals, automobiles, electrical and non-electrical machinery and computers
(Gereffi and Wyman, 1990). Output was destined mainly for local markets, although
beginning in the 1970s more attention went to manufactured exports to offset the
costly import bills associated with ISI deepening. Buyer-driven value chains, by contrast,
were virtually ignored in most of Latin America until recently, because the transnational
firms that established ISI were primarily interested in their local markets, not exports.
This allowed exporters in the East Asian economies that pursued EOI to gain the
lion’s share of US and European markets for the profitable consumer goods supplied
only via buyer-driven chains. With the onset of the Caribbean Basin Initiative in the
mid-1980s and the North American Free Trade Agreement in 1994, Mexico, Central
America and the Caribbean have become major players in US-oriented buyer-driven
chains such as apparel (Gereffi et al., 2002).
Economic globalisation is a kaleidoscopic fragmentation of production processes
and their geographic relocation on a global scale in ways that slice through national
boundaries (Dicken, 1998). Core corporations are shifting from high-volume to high-
value production. Instead of a pyramid, where power concentrates in the headquarters
of transnational firms with vertical chains of command, global production networks
today are webs of independent yet interconnected enterprises. Core firms act as strategic
brokers at the centres of the webs, controlling critical information, skills and resources
needed for the overall global network to function efficiently (Reich, 1991). For
countries and firms to succeed in today’s international economy, they need to position
themselves strategically within these global networks and develop strategies for gaining
access to the lead firms in order to improve their positions.
The evolution of the organisational capabilities of leading firms in the global
economy drives the emergence of new forms of value-chain governance. This
organisational perspective is quite distinct from both the emphasis in neo-classical
economics on pure markets as the key determinant of economic progress and that in
political science on the role of the state in shaping national competitive advantage.
While competitive markets and effective states are clearly important institutional
features of successful modern economies, the global value-chain perspective highlights
a different dimension frequently ignored by these other approaches. It reveals the
shifting bases of power exercised by lead firms in global industries and the ways in
which the governance structures of these industries shape the creation of markets as
well as national development outcomes.

22
 Producer-Driven Chains

Direct foreign investment by transnational corporations was central to the

evolution of producer-driven value chains, given that these companies usually
established international production networks to access raw materials and new overseas
markets. Extending the multidivisional corporate structures pioneered by large US
enterprises to tap the newly emerging American national market (Chandler, 1962),
transnational firms in natural resource sectors such as oil, mining and agriculture set
up international production networks throughout the world to gain access to vital and
profitable raw materials. Breakthroughs in transportation and communication
technologies (e.g. shipping, telegraph and telephone) made integrated production
networks possible (Vernon, 1971), although their ultimate benefits for national
development remain the subject of intense controversy (Barnet and Cavanaugh, 1994).
In the 1950s and 1960s, transnational corporations in the consumer durable and
capital goods manufacturing sectors began to set up their own international production
networks to penetrate overseas markets, especially in Latin America and Asia, which
were regulated by national ISI policies (Gereffi and Wyman, 1990). These companies
had access to the capital, technology and managerial resources essential for the
development of new industries overseas. Because of their emphasis on locally owned
subsidiaries, they had substantial control over the backward and forward linkages in
the entire value chains of which they were parts.

Buyer-Driven Chains

Beginning in the late 1960s, direct foreign investment took a new tack. It
supplemented its resource-seeking and market-seeking motives for globalisation with
a global search for cheap labour. This “new international division of labour” (Fröbel
et al., 1981) relied on further improvements in transport and communication
technologies to slice up the value chain so that the most labour-intensive stages of the
production process could be spatially relocated to areas with the most abundant and
productive low-cost labour. This strategy of firms coincided with the shift by developing
countries from ISI to EOI, facilitated initially by the growth of export-processing
zones in many parts of the developing world (Grunwald and Flamm, 1985).
Traditional accounts of the so-called new international division of labour do not
go far enough, however. First, the transnational manufacturers involved in export-
processing zones typically come from different industries than those involved in the
producer-driven value chains associated with ISI. Export-processing zones tend to
attract “light industries” (e.g. apparel, footwear, consumer electronics, toys), with
relatively low barriers to entry. Second, the production-oriented frameworks — whether
espoused by neo-classical economists, Marxist scholars, or the World Bank — miss
the role of commercial capital in the globalisation process. This is the major feature
of the distinction between producer-driven and buyer-driven global commodity chains
(Gereffi, 1994, 1999). Third, global sourcing in buyer-driven chains is driven by

23
intense competition among different types of developed-country retailers and
marketers2 who feel compelled to mimic each other’s moves in two ways: in growing
offshore sourcing networks and in using brands as sources of market power.
Particularly noteworthy in the shift to both buyer-driven and internet-oriented
value chains is the growing importance of global brands, which can be created without
proprietary links to specific manufacturers or distribution channels. “Brands are the
information — whether real or imagined, intellectual or emotional — that consumers
associate with a product” (Evans and Wurster, 2000, p. 11). For consumers, brand
knowledge is simply a high-richness/low-reach stock of information that comes from
advertising, reputation and especially prior experience, in contrast with the high reach
and low richness of classic markets. For companies, brands are a way to resist
commodification when value chains deconstruct. Sellers use brands to lock in customer
relationships3 and to compete when reach (choice) goes up. Thus, building brand
awareness is a fundamental challenge and a major source of market power for firms in
both buyer-driven and internet-oriented value chains.
One of the main sources of organisational innovation in the shift from producer-
driven to buyer-driven value chains derives from the disarticulation of brands and the
production process. Originally, leading manufacturers developed brands to differentiate
their products from those made by competitors (e.g. General Electric light bulbs,
Levi’s jeans, Kodak film). Both retailers and marketers then decided to tap into the
profitability enjoyed by branded manufacturers. To distinguish themselves from larger
and more diversified department stores, emergent speciality retailers sold only one
kind of product under the store’s own brand name (e.g. The Limited, Victoria’s Secret,
The Gap, Benetton). Marketers like Nike, Reebok, Liz Claiborne, and Ralph Lauren
took the branding concept one step farther by eschewing both factories and actual
stores; their profitability derived solely from elaborate promotional schemes based on
carefully crafted “lifestyle brands” associated with their products. Department stores
fought back with private-label merchandise — i.e. stores’ brands that competed with
the top national brands on price but nonetheless had better quality and style than other
products4. Today, brands are even dissociated from specific products and linked to the
internet infomediaries that channel information to web-based consumers (America
Online, for example, is one of the best known brands, even though it is only a web-
based navigator).
Table 1.1 summarises a number of the evolutionary shifts alluded to above.
Whereas producer-driven global value chains are characterised by vertical integration
by transnational corporations based on ownership and control, buyer-driven chains
highlight the global sourcing networks established by retailers and marketers that rely
heavily on sophisticated logistics and performance trust among numerous contractors.
Table 1.1 uses the dates when leading US retailers and marketers were founded, or
went public, to trace the sequential entry into the global sourcing game of large retailers,
pure marketers and speciality retailers in the 1970s, and private-label (store brand)
programmes in the 1980s. In the 1990s, the information revolution motivated the

24
direction and pace of organisational innovation in the current shift to internet-oriented
value chains based on virtual integration and an explosion in connectivity due to the
open and almost cost-free exchange of a widening universe of rich information.

Electronic Commerce and the Reorientation of Value Chains

The economic transformation at the turn of the 21st century, driven by the often
spectacular development and diffusion of modern electronics-based information
technology, has a variety of names — innovation economy, knowledge economy,
network economy, digital economy and E-conomy (Cohen et al., 2000). Yet electronic
commerce is not simply a creature of technology. It also involves profound changes in
business organisation, market structures, government regulation and human experience.
The internet already is beginning to have a significant impact on the structure and
competitive dynamics of global value chains5.
The two most important types of e-commerce occur in business-to-consumer
(B2C) and business-to-business (B2B) markets6. The B2C market transfers goods and
services to individual consumers (a retail model), whereas B2B entails procurement,
logistics and administrative processes between firms (a supply-chain model). Electronic
commerce is growing so rapidly that estimates of the magnitude of these two markets
vary widely. For example, the Boston Consulting Group estimates total on-line retail
sales of $34.2 billion in 1999. Forrester Research calculates on-line sales of $20 billion
in the B2C market in 1999, but expects that figure to grow to $184 billion by 2004
(U.S. Department of Commerce, 2000, pp. 42-43; The Economist, 2000a, pp. 9-10).
B2B transactions dwarf on-line retail sales, however. They account for as much as
80 per cent of all e-commerce. According to the Gartner Group, a Connecticut-based
market research firm, the worldwide B2B market will grow from $145 billion in
1999 to $401 billion in 2000 and $7.3 trillion by 2004 (i.e. 7 per cent of the forecast
$105 trillion in worldwide sales transactions) (Standard & Poor’s, 2001, p. 2).
The internet has the potential to transform both buyer-driven and producer-
driven chains because of two fundamental factors: i) its ability to create markets on
scales and with levels of efficiency not previously possible7; and ii) a radical “pull”
business strategy that substitutes information for inventory and ships products only
when there is real demand from customers. The shift from manufacturer “push” to
consumer “pull” appears to be a long-term trend in many industries. It places a premium
on a “build-to-order” business model and reflects a focus on consumer satisfaction
and convenience (see the discussions of AOL, Amazon.com, and Dell in Gereffi, 2001).
The “pull” strategy in supply-chain management is embodied in popular business-
school concepts such as mass customisation (Pine, 1992), lean production (Womack
et al., 1990) and lean retailing (Abernathy et al., 1999).

25
 Table 1.1. The Historical and Institutional Origins of Changing Governance Structures in Global Value Chains
a
Governance Structure of Leading Industries Main Drivers Form and Dominant Institutional and Corporate and National Pioneers
Global Value Chains and Timing Principles of Value Chain Organisational Innovations
Integration
Producer-Driven Chains Natural resources: late Transnational Vertical integration Vertically integrated Oil companies (1870s onward)
19th and early 20th manufacturers (ownership and control) TNCs with international Mining (early 20th century)
centuries production networks Agribusiness (early 20th century)
Capital goods and Fordism (1920s onward)
consumer durables: Mass production
1950s and 1960s Japanese TNCs (Toyota, early
Lean production 1960s on)
Buyer-Driven Chains Consumer Retailers and Network integration Growth of export Mexico, the Philippines, Chinese
nondurables: marketers (logistics and trust) processing zones Taipei, South Korea, etc. (mid-
1970s & 1980s 1960s onward)
Global sourcing by Sears, Kmart, Montgomery Ward,
retailers JC Penney (early 1970s onward)
Liz Claiborne (1976). Nike (1976),
Rise of pure marketers Reebok (1979)
The Limited (1969), The Gap
26

Rise of specialty (1976)

retailers JC Penney, Sears, Wal-Mart, Kmart
Growth of private labels (mid-1980s onward)
(store brands) Wal-Mart, JC Penney, Dillard’s,
Lean retailing Federated (late 1980s onward)

Internet-Oriented Chains Services (B2C)
(emerging) – online retailing
– online brokerage
Intermediates (B2B)
– autos (Covisint)
– computers
Internet infomediaries Virtual integration Rise of e-commerce Amazon.com (1997)
(B2C market) and (information and access)
some established Mass customisation Dell (1988), Gateway (1993)
manufacturers (B2B
market) Disintermediation: E*Trade (1992), Schwab (1996)
– Direct sales (skip
retailers)
– Online services (e.g.
brokerage)

1990s & 2000s New internet navigators AOL (1992), Yahoo! (1996),
Excite@Home (1999)
Note: a) Specific dates indicate when companies were founded, went public (The Limited, The Gap, Dell, Gateway, Amazon, AOL, Yahoo!), or became established as US firms (Nike,
Reebok). Decades are used for the onset of trends.
 The B2C Market

One of the early changes attributed to the internet is the emergence of a new
breed of “infomediaries” — companies that turn on-line access to customers, and
especially detailed information about their purchasing habits, into a highly valued
asset. Although most infomediaries in B2C transactions currently represent the interests
of consumers trying to get the most out of the web, internet navigators are also affiliated
with producers, sellers and traditional value-chain intermediaries. This infomediary
model is based on oligpolistic competition, in which dominant infomediaries like
AOL or Yahoo! control portals8 and other strategic entry points to the internet. They
further leverage their power by becoming more integrated across the internet
organisational chain through mergers, acquisitions and strategic alliances.
Table 1.2 shows the internet’s organisational chain, composed of the main firms
that make internet transactions possible. The internet is an interconnected global network
of smaller networks that links millions of computers through thousands of servers. It
is built on a complex hardware infrastructure of internet equipment providers, computer
makers and component suppliers, integrated by software and services. Companies
such as Cisco Systems, Nortel Networks, and Lucent Technologies dominate the market
for internet equipment, such as routers and remote access concentrators. The main
customers for the internet are businesses (B2B markets) and individual consumers
(B2C markets), with B2B markets currently dwarfing B2C transactions in size.
Computer makers are an integral part of the internet chain because most businesses
and individuals hook up to it with personal computers. Corporate clients look for
“single solutions” to meet their increasingly complex computing needs, and thus the
major computer companies have shifted their focus to three main areas of growth:
servers, storage and services. The expected demand for global technology services, by
far the largest of the three, will nearly double from $359 billion in 1999 to over
$700 billion by 2004 (Standard & Poor’s, 2000e, pp. 3-6). Firms such as Oracle,
Ariba, Commerce One and i2 Technologies that develop software for on-line
transactions are becoming key players in the rapidly emerging B2B marketplace. Other
important links in the internet organisational chain are:
— web browsers (browser software permits on-line navigation by allowing users to
view the text and graphics located on internet web sites);
— internet service providers (ISPs offer basic, flat-rate internet access to customers);
— internet content providers (ICPs use mostly original material to create internet
destinations where people go for information, entertainment, or commerce).
In each segment of the internet organisational chain, the leading companies
have dominant market shares. Cisco controls more than three-quarters of the global
market for internet routers and switches. The top four personal computer vendors
(Compaq, Dell, IBM and Hewlett-Packard) account for nearly 40 per cent of unit
shipments worldwide. Microsoft controls about 90 per cent of personal computer

27

Internet Equipment Personal Computer (PC)
Suppliers1 Manufacturers and
Component Suppliers2
Table 1. 2. The Internet’s Organisation Chain
4 7
PC and E-Business Web Browsers Internet Service Internet Content Customers
Software3 Providers (ISPs)5 Providers (ICPs)6

Cisco Systems PCs: Compaq. Dell, PCs: Microsoft, Microsoft AOL AOL Businesses (B2B)
Lucent Technologies Hewlett-Packard, Apple. Netscape/AOL Microsoft Microsoft – Covisint (Autos)
Nortel Networks IBM. Servers: Unix, Linux. AT&T Yahoo! Consumers (B2C)
Sun Microsystems Microprocessors: E-Business: Oracle, Lycos – Amazon.com
Intel, AMD. Ariba, Commerce Excite – Dell Computers
Disk Drives: One, SAP
Seagate, Quantum
Notes and Sources:
1. Cisco controls more than 75 per cent of the global market for internet routers and switches (Hoover’s Company Profiles, 2000). Sun Microsystems is the main supplier of the Unix
28

server market with a 28 per cent market share in 1999, followed by Hewlett-Packard (23 per cent) and IBM (18 per cent) (Standard & Poor’s, 2000e, p. 4).
2. Compaq had 13.1 per cent of third-quarter 2000 worldwide PC shipments, followed by Dell (11.5 per cent), Hewlett-Packard (7.8 per cent) and IBM (7.4 per cent). In the US market,
however, Dell (20.2 per cent) pulled ahead of Compaq (16.5 per cent) as the leading PC vendor in third-quarter 2000 sales. More than 80 per cent of the world’s personal computers
include an Intel microprocessor (Standard & Poor’s, 2000e, pp. 2-3, 18).
3. Microsoft’s Windows operating system software is used in an estimated 85–90 per cent of PCs worldwide (Standard & Poor’s, 2000e, p. 12). Unix and Linux power most web servers
with proprietary and open-source software, respectively, while Oracle, Ariba, Commerce One, and Germany’s SAP are among the largest providers of software for e-business
transactions.
4. In June 2000, Microsoft’s Internet Explorer had 86 per cent of the web browser market and Netscape’s Navigator (now a part of AOL) the remaining 14 per cent (Standard & Poor’s,
2000d, p. 9).
5. America Online had 43 per cent of the ISP market in 1999, followed by Microsoft (6 per cent), and AT&T WorldNet (5 per cent) (Standard & Poor’s, 2000c, p. 3). This was prior to
the merger of AOL with the cable giant Time Warner Inc., announced in January 2000 and finally approved one year later.
6. The most visited web properties in July 2000 (i.e. the percentage of web-active individuals who visited a site at least once during the month) were the AOL network (worldwide web
and proprietary — 78 per cent); Microsoft sites (63 per cent); Yahoo sites (61 per cent); Lycos (40 per cent); and Excite (34 per cent) (Standard & Poor’s, 2000d, p. 4).
7. Global B2B e-commerce is predicted to reach $4 trillion by 2003, a market 10 times bigger than the $400 billion forecast for B2C online sales to consumers in 2003 (The Economist,
2000b, p. 11).
operating systems, and two-thirds of the web browser market. America Online (AOL)
had 43 per cent of the internet service provider market in 1999 and more subscribers
than the next 20 ISPs combined. AOL, Yahoo! and Microsoft sit atop the internet
content provider market as well. (See Table 1.2 for references.)

The B2B Market

The automotive industry leads in B2B e-commerce. It contains the world’s largest
on-line marketplace to date. Covisint is a newly formed joint venture that combines the
purchasing activities of General Motors, Ford, Daimler–Chrysler, Renault, Nissan and
their suppliers. Initially announced in February 2000 as a joint electronic supply agreement
among General Motors, Ford and Daimler-Chrysler, Covisint subsequently added Renault/
Nissan, and in October 2000, following US Federal Trade Commission clearance, the
first on-line auctions took place (Standard & Poor’s, 2000b, p. 5). The scope of the
venture is staggering. In 1999, General Motors’ total automotive purchases were
approximately $87 billion, Ford’s were $85 billion, and Daimler–Chrysler’s were
$80 billion. Each does business with about 30 000 suppliers. Estimated annual transactions
on the exchange will exceed $240 billion, and the venture is expected to shave billions of
dollars off procurement costs9. Commerce One and Oracle have been brought in as
technology partners to help develop on-line software for the auto parts exchange.
Covisint promises lower prices, faster transaction turnarounds and other
efficiencies, but many suppliers fear they could become losers because lower prices
for buyers will mean lower margins for sellers. While Covisint strives for unprecedented
collaboration among the world’s leading automakers, other equally significant changes
are occurring. The tight vertical structures that used to bind most parts suppliers to
particular car manufacturers are loosening. The large, technologically sophisticated
global suppliers (such as Bosch, Denso, Johnson Controls, Lear Corporation, TRW
and Magna) are gaining strength, as they become preferred partners in all the major
automakers’ supply chains (Sturgeon and Florida, 1999). These developments, together
with the emergent megadealers in automotive retailing10, could lead to substantial
realignments in the relative power and profitability of major segments in the automotive
value chain.

Three Scenarios for the Internet’s Impact on Global Value Chains

The internet remains in its early stages of development, but its impact on global
value chains is already evident. While it may be premature to try to identify lasting
changes on producer-driven and buyer-driven chains, three possible and not mutually
exclusive scenarios are emerging. In the first, the internet will lead to the formation
of new infomediary-based value chains, which implies a different set of organisational
drivers. Although some spectacularly successful e-commerce ventures appeared in the
late 1990s, the B2C market is still too small and volatile to establish a radically
distinct and durable governance structure.

29
 In the second scenario, the internet significantly extends the logic of buyer-
driven chains as both information and power continue to shift inexorably from
producers and retailers to consumers. Rather than serving as an alternative to buyer-
driven chains, the internet intensifies a shift that makes all industries more “buyer
driven” — in the sense that new consumer-oriented competitors undermine the power
of manufacturers, retailers, and marketers that do not take advantage of the internet’s
ability to facilitate mass customisation.
The third scenario sees the impact of the internet in both B2B and B2C transactions
captured and integrated into the business practices of leading manufacturers, retailers,
and marketers. Pitting the so-called new economy against the old economy completely
misses the point because the internet’s major impact will be to improve the productivity
of all parts of the economy, especially the old-economy firms. Established leaders in
both producer-driven and buyer-driven chains are proving surprisingly adept at
incorporating e-commerce in their business strategies (in popular jargon, these companies
are moving from “bricks and mortar” to “clicks and mortar”). Thus, the biggest and
most powerful companies co-opt and internalise the internet, and they force their
rivals and suppliers alike to bear the costs of adapting to new information technologies.
While there is evidence to support all three scenarios, the third model currently
seems dominant. Nonetheless, lead firms in major industries are adopting quite different
strategies for key supply-chain issues such as vertical integration, outsourcing and
globalisation. The impact of the internet on these business structures remains an
open question.

30
 Notes

1. Although the concepts of global value chains, global commodity chains and global
supply chains are closely related, the term “global value chains” will be used in this
paper to emphasise the flows of information, resources, goods and services along
the full range of activities and organisations in the supply chain. For a more detailed
discussion of the origins and use of these concepts, see Gereffi (1994), Kaplinsky
(2000), and Raikes et al. (2000).
2. In the original distinction between producer-driven and buyer-driven commodity
chains (Gereffi, 1994), the term “buyer” was used in an organisational sense to refer
to retailers and marketers (in their pure form, marketers build and commercialise
their own brand names but own neither factories nor stores). These organisational
buyers are to be distinguished from the actual individual consumers who are the
targets of much contemporary e-commerce.
3. Brand affiliation is surprisingly stable. Of the 25 top-selling consumer goods brands in
1960, 16 of them are still among the top 25 today (Evans and Wurster, 2000, p. 150).
4. This trend is most visible in apparel products (see Standard & Poor’s, 2000a). In blue
jeans, for example, JC Penney’s Arizona jeans and Sears’ Canyon River Blues go head-
to-head with upscale jeans sold by Tommy Hilfiger, Calvin Klein and Donna Karan.
5. The material in the remainder of the paper is summarised from the more detailed
discussion in Gereffi (2001).
6. If one were to complete the e-commerce matrix, the consumer-to-business (C2B)
market would be represented by Priceline.com, the most popular of several reverse-
auction sites, while the consumer-to-consumer (C2C) segment includes consumers’
auctions, epitomised by the auction site eBay.com. See The Economist (2000a) for
a fuller analysis of the e-commerce matrix.
7. A couple of familiar companies provide good illustrations of the extensive reach
provided by the internet. Amazon.com, one of the first electronic retailers on the
web, has no physical stores but offers an electronic list of 3 million books, 20 times
larger than the holdings of Barnes & Noble, the largest chain bookstore. Dell’s
internet site offers over 10 million computer configurations (Evans and
Wurster, 2000, pp. 61-62, 111).
8. Portals are web sites designed to be an internet user’s initial entry point for exploring
the web. They typically generate revenues by renting out advertising space.
9. The on-line exchange is expected to yield savings of $2 000-$3 000 on a $19 000
vehicle (Covisint, 2001).
10. AutoNation is the largest car dealer in the United States. It had sales of
approximately $1 billion (about 46 000 vehicles) via the internet in 1999 (Standard
& Poor’s, 2000b, p. 5).

31
 Bibliography

ABERNATHY, F.H., J.T. DUNLOP, J.H. HAMMOND and D. WEIL (1999), A Stitch in Time: Lean
Retailing and the Transformation of Manufacturing — Lessons from the Apparel
and Textile Industries. Oxford University Press, New York, NY.
BARNET, R.J. and J. CAVANAUGH (1994), Global Dreams: Imperial Corporations and the New
World Order, Simon & Schuster, New York, NY.
B ARNET, R.J. and R.E. MÜLLER (1974), Global Reach: The Power of the Multinational
Corporations, Simon & Schuster, New York, NY.
CHANDLER, A.D., JR. (1962), Strategy and Structure: Chapters in the History of the American
Industrial Enterprise, MIT Press, Cambridge, MA.
COHEN, S.S., J. BRADFORD DELONG and J. ZYSMAN (2000), “Tools for Thought: What is New
and Important about the ‘E-conomy’?”, BRIE Working Paper No. 138, Berkeley
Roundtable on the International Economy, Berkeley, CA.
COVISINT (2001), General FAQ’s about Covisint [online]. http://www.covisint.com/info/
faq_gen.shtml (down-loaded on Feb. 17, 2001).
DICKEN, P. (1998), Global Shift: Transforming the World Economy, 3rd edition, Guilford
Press, New York, NY.
EVANS, P. and T.S. WURSTER (2000), Blown to Bits: How the New Economics of Information
Transforms Strategy, Harvard Business School Press, Boston, MA.
FRÖBEL, F., J. HEINRICHS and O. KREYE (1981), The New International Division of Labor,
Cambridge University Press, New York, NY.
GEREFFI, G. (2001), “Shifting Governance Structures in Global Commodity Chains, with
Special Reference to the Internet”, American Behavioral Scientist, Vol. 44, No. 10.
GEREFFI, G. (1999), “International Trade and Industrial Upgrading in the Apparel Commodity
Chain”, Journal of International Economics, Vol. 48, No. 1.
GEREFFI, G. (1994), “The Organization of Buyer-Driven Global Commodity Chains: How
U.S. Retailers Shape Overseas Production Networks”, in G EREFFI , G. and
M. KORZENIEWICZ (1994).
GEREFFI, G. and R. KAPLINSKY (eds.) (2001), The Value of Value Chains: Spreading the
Gains from Globalisation, IDS Bulletin, Vol. 32, No. 3.

32
GEREFFI, G. and M. KORZENIEWICZ (eds.) (1994), Commodity Chains and Global Capitalism,
Praeger, Westport, CT.
GEREFFI, G., D. SPENER and J. BAIR (eds.) (2002), Free Trade and Uneven Development: The
North American Apparel Industry after NAFTA, Temple University Press,
Philadelphia, PA.
G EREFFI , G. and D.L. W YMAN (eds.) (1990), Manufacturing Miracles: Paths of
Industrialisation in Latin America and East Asia, Princeton University Press,
Princeton, NJ.
GORE, C. (2000), “The Rise and Fall of the Washington Consensus as a Paradigm for
Developing Countries”, World Development, Vol. 28, No. 5.
GRUNWALD, J. and K. FLAMM (1985), The Global Factory: Foreign Assembly in International
Trade, The Brookings Institution, Washington, D.C.
HOOVER’S ONLINE (2000), Company Profiles [online]. http://www,hoovers.com.
HUMPHREY, J. and H. SCHMITZ (2001), “Governance in Global Value Chains”, IDS Bulletin,
Vol. 32, No. 3.
KAPLINSKY, R. (2000), “Spreading the Gains from Globalisation: What Can Be Learned
from Value Chain Analysis?”, Journal of Development Studies, Vol. 37, No. 2.
PINE, B. J. (1992), Mass Customization: The New Frontier in Business Competition, Harvard
Business School Press, Boston, MA.
RAIKES, P., M.F. JENSEN and S. PONTE (2000), “Global Commodity Chain Analysis and the
French Filière Approach: Comparison and Critique”, Economy and Society, Vol. 29,
No. 3.
REICH, R.B. (1991), The Work of Nations: Preparing Ourselves for 21st-Century Capitalism,
Alfred A. Knopf, New York, NY.
STANDARD & POOR’S (2001), Industry Surveys, “Computers: Commercial Services”, 25 January.
STANDARD & POOR’S (2000a), Industry Surveys, “Apparel & Footwear”, 5 October.
STANDARD & POOR’S (2000b), Industry Surveys, “Autos & Auto Parts”, 28 December.
STANDARD & POOR’ S (2000c), Industry Surveys, “Computers: Consumer Services & the
Internet”, 16 March.
STANDARD & POOR’S (2000d), Industry Surveys, “Computers: Consumer Services & the
Internet”, 28 September.
STANDARD & POOR’S (2000e), Industry Surveys, “Computers: Hardware”, 14 December.
STURGEON, T. and R. FLORIDA (1999), “The World that Changed the Machine: Globalisation
and Jobs in the Automotive Industry”, unpublished final report to the Alfred P. Sloan
Foundation, 5 May.
The Economist (2000a), “Shopping Around the Web: A Survey of E-Commerce”,
26 February.
The Economist (2000b), “Untangling E-conomics”, 23 September.

33
US DEPARTMENT OF COMMERCE (2000), U.S. Industry & Trade Outlook 2000, McGraw Hill/
U.S. Department of Commerce, New York, NY.
VERNON, R. (1971), Sovereignty at Bay: The Multinational Spread of U.S. Enterprises,
Basic Books, New York, NY.
WOMACK, J.P., D.T. JONES and D. ROOS (1990), The Machine That Changed the World, Rawson
Associates, New York, NY.
WORLD BANK (1993), The East Asian Miracle, Oxford University Press, New York, NY.

34
 Chapter 2

E-Commerce for Development:

A General Framework
Patrizia Fariselli

Introduction and Summary

Change and Models

Policies to overcome the digital divide or prevent its emergence in developing

countries need grounding in a general view of the impact of the digital economy on
development and growth — a view not yet clear or formalised into robust models.
OECD work to analyse the contribution of information and communications
technologies (ICTs) to macroeconomic performance requires the reconstruction of
indicators and accounting frameworks. The interplay between the dynamics of socio-
economic change — induced and/or altered by pervasive ICT infrastructures, services,
and goods — and those of research and policy gives expectations, case studies and
short-term analyses a prevailing and sometimes misleading role. Potential network
effects make the measurement of impacts from application of new ICTs that much
more difficult but at the same time more important.
Early attention to electronic commerce tended to focus on business-to-consumer
(B2C) trading. Forecasts of phenomenal B2C growth proved exaggerated, sometimes
in the extreme, and recent prognoses are more sober. More recently, much attention
has gone to business-to-business (B2B) e-commerce and its potential, but once more
the integration of B2B models within the supply chains of the so- called “old” economy
has been slower than expected. Euphoric expectations of the contribution of electronic
commerce to corporate profit growth produced over five years a gigantic investment
bubble in dotcom start-up companies that has since been deflated. Looking at the new
economy in this light raises the question of what if anything is really new. The change

35
is not in investors’ behaviour, which has always swung between optimism and
pessimism, but apparently in how international capital markets function. It derives
primarily but not exclusively from technological change — the pervasiveness of ICTs
and the way they make the information and knowledge content of transactions
transparent. The major impact is on the configuration of the networks (ICT
infrastructure, financial and business organisations and research, education and policy
networks). The possibilities both of becoming interactive and global and of decomposing
and recomposing operations and supply chains strain the stability of established
networks. The application of ICTs makes a network’s boundaries and configuration
extraordinarily flexible. ICTs perform an enabling function to the extent that they
provide methods, tools and solutions to address increasingly complex computational
and system organisation and implementation requirements.

Access to Networks and Access to Knowledge

The digital economy’s boundaries go beyond the trading of intangible goods,

typical of electronic commerce at its early stage, to encompass all business and
community relations. The notions of the “knowledge economy” and the “net economy”
become more appropriate than that of electronic commerce. The link between ICT
and development is represented basically by access to networks — the global ICT
infrastructure, business organisations, research and education communities and policy
institutions — and by access to information and knowledge. Information and knowledge
can be valuable resources that command a market price but they also have important
public-good properties, with an inherent tension between their private and public
aspects. Technologies of digital reproduction also make it ever easier to share information
over the internet, causing certain industries — notably the music industry — to redouble
efforts at strict copyright enforcement. This is a crucial economic and policy issue
concerning the relationship between ICT and development. Beside the debate about
IPRs on information and knowledge products currently in the market, the generation
of new information and knowledge is influenced by the accessibility of those products
as a means of producing new knowledge.
Countries and regions lagging behind in ICT diffusion can take advantage of
business and policy experience accumulated in the developed world over the last several
years, but that does not necessarily imply that they should replicate the same paths or
sequences. The rich countries are experiencing a set of innovations — in technologies,
in markets, in business organisations, in public administration and in socio-cultural
environments — which open up multiple scenarios for the future. The end result in
economic growth and social welfare is an open question. A digital divide between the
haves and have-nots is perceived as a risk both between and within countries.
The components influencing the take-up of ICT and access to the digital economy
may be grouped into the four categories listed below. How and how intensely these
components are associated contribute to the occurrence, deepening or reduction of the
digital divide.

36
 — Physical Network ® ICT Infrastructure and Systems
— Economic Network ® Market and Business Organisations
— Resources ® Information and Knowledge; Financial Capital
— Policy ® Support and Regulatory Frameworks
The digital divide is evident in the short run from a comparison of ICT market
indicators across countries or regions, and the reduction of transaction costs in business
organisations based on e-commerce is measurable with currently available accountancy
tools. Yet assessment of the contribution of digital technologies to development and
growth remains a work in progress. Besides economic variables, it must also look at
policies, because the role played by policy institutions in technology and infrastructure
access deeply influences the divide and development prospects. Barriers to IC networks,
exclusion from digital marketplaces or communities and exclusion from the generation
and sharing of knowledge and innovation processes will all likely produce market
fragmentation, loss of resources, reduction of pluralism and a slowdown of technical
progress and innovation, in both developed and developing countries. Only policy
actions can prevent these effects.
Developing countries or areas cannot reasonably be expected to follow entirely
the same paths as the most developed ones in the transition to the digital economy.
They can leverage the same catalytic components, particularly the policy components,
but they cannot rely to the same extent on the prospect of reducing transaction costs in
markets and organisations. The priority in the weakest economies is to build capacity
and critical mass rather than reduce the use of resources. Except where resources get
incorporated into supply chains under the control of large transnational corporations
— which amounts to absorbing e-commerce practices passively — the problem usually
involves just having in place a sufficient critical mass of local businesses to make
electronic interrelations possible and effective.
On the other hand, the trend towards technological convergence offers
underdeveloped areas the opportunity to leapfrog some phases of the development of
telecommunications infrastructure in the industrialised world over the last century.
Mobile telephony/internet and satellite communication networks can easily be set up
where fixed networks in place are not capillary. New technology applications and a
liberalised market for carriers allow bypassing the hurdles raised by bureaucratic
incumbents — the old PSTNs — and facilitate less expensive mass access to the modern
communication networks. Yet the pricing practices of broadband service and satellite
network suppliers still represent a bottleneck to access to the global communications
infrastructure. Public investment in telecommunications infrastructure becomes crucial,
as it has been and still is in the most industrialised countries. There, a universal
service philosophy and low-cost access to the internet have been supported by state
administrations to the joint advantage of citizens, public and private carriers and the
ICT industry.

37
 Local Generation of Knowledge

Besides access to the physical infrastructure, access to knowledge is the major

opportunity and challenge in the digital economy. The erosion of time, space and cost
barriers, and the decentralised network model, make ICT extremely powerful in
stimulating or accelerating growth. They allow knowledge resources to break the
chains imposed by old and rigid business organisation models and favour creativity in
exploring new application fields. The creation and trading of knowledge have become
key objectives of economic activity everywhere. As global as the knowledge market
is, however, knowledge has an essentially local dimension. Part of the reason for a
digital divide, besides limited access to the physical network, is the limited quantity
and quality of knowledge under local control. Good practices in electronic commerce
from the most developed countries and markets must be adapted to local contextual
requirements if they are to generate local development and growth and avoid the
creation of islands connected to corporations based elsewhere but disconnected from
the local context.
By decomposing and recomposing network components through the isolation
and transfer of information and knowledge, ICT and the internet contribute to blurring
geo-economic borders and shifting comparative advantage. While certain high-tech
industries and activities remain highly concentrated in a few noteworthy agglomerations
in the developed world, important new entrants include such traditionally marginal
areas as Sweden, Finland, Brazil, India, Israel and perhaps South Africa. The same
holds at the sub-national or regional level. The well established practice of outsourcing
to India segments of data processing, software development and internet-related services
by large US or Europe-based corporations is evolving toward the establishment of
local business units (delocation) and the creation of new local enterprises able to
compete in the global market. India’s successes show the boost possible from a
combination of policy support and private investment in physical infrastructure, business
creation, human resources and education.

From Divide to Development

The extension of ICT and internet networks and access to them are preconditions
for any form of electronic commerce and the uptake of the digital economy anywhere.
The penetration of ICT and the cost of access are reshaping the global map. In the
LDCs, the picture is differentiated.
Pro-active policies can make the difference. Public Administration (PA) can
play a strategic role in piloting the transition to the information society and to the
digital economy. Internet applications within government include PA-to-Citizens, PA-
to-Business, and PA-to-PA. The internet can serve as a development tool that permits
government simultaneously to address the needs of internal and international integration
and faster and wider diffusion of public services such as education and training, health
care and assistance to business creation and management. Systemic issues related to

38
electronic commerce, whether legal (contracts, security, confidence, IPR), financial
(e-payment, insurance, taxation), or technical (interoperability of systems and
standards), have simultaneously a national and an international dimension. Adopting
e-government frameworks and procedures enables LDC governments to participate
effectively in the global forums setting international standards and the rules for
international trade in the digital economy. PA can also play a significant role in
supporting the development of economic institutions and of the knowledge resources
needed to enable local economies to compete in global markets. Policies to build local
capacity could be associated with measures (e.g. on-line procurement) to enhance
effective local demand for internet services and to encourage the production of local
content and local provision of internet-related services and applications.

What is E-Commerce?

Business-to-Consumers

A tide of new terminology floods current language, spreading from specialist

disciplines through the mass communications media. Symbols, prefixes and acronyms
such as @, e-, and B2B have become signposts to identify a largely uncharted territory.
Indicators and models are still in process, so that examples often substitute for definitions
and have an imprint on them. Widely accepted definitions of electronic commerce
delivered by the European Commission (1997) and the OECD (1998) reflected a then
prevailing view of electronic commerce as buying and selling consumer goods over
the internet, particularly software, music, books, travel, culture, hobbies and advertising.
The services to develop in that early stage were mostly electronic catalogues, electronic
malls, and anything connected with the design, operation and maintenance of
commercial web sites and related data bases.
At first, the global and decentralised internet unveiled the global market as a
place that consumers and suppliers, individuals or companies, could access directly,
bypassing intermediaries, taxes and bureaucratic hurdles. In a way, the nature of the
interaction tool (internet) was transferred to the object of the interaction itself
(commercial transactions). Yet the models required to implement electronic trading
over the internet efficiently and effectively also include the (re)setting of manifold
components involving technology and economics, business management and regulations,
content and behaviour. All are usually listed as “barriers” to the uptake of electronic
commerce. Local and international policies get launched to support and enhance the
“readiness” for electronic commerce and its governance around the globe. In fact, the
possibility of entering the global marketplace in a one-to-one relationship (consumer-
to-producer and vice versa) endangers the role of traditional intermediaries, especially
in retailing and wholesaling. It also creates, however, a wide population of new on-
line intermediaries, as providers of equipment, network access and services, web-
related services, data warehouses, business intelligence, content management, financial
tools and services for on-line transactions, marketing, advertising and consulting.

39
 Empirical studies provide mixed evidence on internet market efficiency properties
(price levels, price elasticities and menu costs) compared with conventional retail
channels for a number of products such as CDs, books, software, used cars and airline
tickets. Price dispersion is never lower in internet markets. It is substantially higher
(between 20 per cent and 33 per cent) for books, CDs and airline tickets (Smith et al.,
1999)1. Many factors contribute to the decision to shop on-line and to make specific
purchases. Price appears to be only one of them, and the lowest price does not necessarily
take priority in customers’ preferences. The cultural and life-style profiles of internet
shoppers and the marketing policies of suppliers both influence consumer behaviour.
Time is a highly valued commodity in internet business2, and the time efficiency that
internet-enabled search offers to consumers may make some willing to pay price
premiums in exchange for time savings. This partly explains the persistence of price
dispersion. Information about consumers’ profiles, tastes and preferences adds value
to electronic transactions exceeding their market prices, because it becomes an input
to the customer-oriented production of the goods, services and advertising delivered
by suppliers. In a way, consumers participate in the process as both consumers and
producers of information incorporated in the supply of goods and services. Research
work on the weightless economy focuses on this interplay (Quah, 1999).

Business-to-Business

A greater appreciation of the importance of business-to-business applications

has come more recently to balance the initial emphasis on e-shopping. Awareness of
the potential impact of internet technologies on business processes behind the front
office has grown apace. Business-to-business transactions represent by far the largest
turnover in e-commerce markets. Since the mid-1960s, the adoption of ICT application
systems has steadily increased in business management, along with the evolution of
computer and component technology (Schiller, 2000). ICTs have become increasingly
essential to the firm and its management as the complexity of the processes and relations
underpinning the enterprise increases. The complexity derives from the web of
interdependencies and the nexus of contracts within which the typical firm is situated.
When internet technologies are coupled with ICTs, electronic commerce allows
enterprises to do more with less (Kalakota and Whinston, 1997) by reducing the costs
of managing the business processes underlying the enterprise’s activities along the
value chain.

E-Business

A number of business models have developed along with ICT-driven business

processes and supply chain integration (Lefebvre and Lefebvre, 2000). The evolution
of e-commerce from one-to-one applications (e-tailing) to one-to-many applications,
such as enterprise and supply chain integration, e-procurement, e-auctions,
e-marketplaces, collaborative commerce and virtual communities, is rendering virtual

40
both the value-added activities of an enterprise and the enterprise itself. It reinforces
the shift from the enterprise-centric vision to the multi-enterprise network or virtual
enterprise3. The key implication is that ICT and internet use can not only reduce costs
but also create value. Value is generated from brokering transactions and matching
orders between companies as well as from the provision of additional services. Those
services can include ASP (application and professional services for integrating
companies’ legacy systems into the e-marketplace), business services to support
transactions (legal and financial, logistics, project and contract management),
background services (market intelligence) and services to set up virtual private
marketplaces for smaller and targeted e-business communities.
In an antidote to the vagueness of most data concerning internet commerce and
particularly business-to-business dealings, EITO 2001 provides a more precise definition
of e-business. E-business is an umbrella term. It encompasses e-commerce and
marketing/information web sites, as well as the concept of E-business services, i.e. the
range of ICT support services associated with the customisation, implementation and
ongoing running, hosting and maintenance of one-to-many e-commerce solutions and
e-marketplaces, plus those associated with marketing/information web sites. The total
European e-business services market was worth Euro 11.4 billion in 2000. The market
for one-to-many e-commerce ICT services was Euro 6.2 billion (55 per cent of the
total), distributed as follows: 45 per cent in implementation services, 41 per cent in
application hosting, 8 per cent in other ongoing support services and 6 per cent in
e-ordering-specific services (EITO, 2001).
The rate of penetration of one-to-many e-commerce applications is relatively
high in the service sector (with finance/insurance reporting an e-ordering rate of 45 per
cent of total orders) compared with manufacturing and the public sector. In manufacturing,
the companies driving the buy-side e-commerce solutions are those with big supply
chains and heavy procurement expenditures, such as automotive, aerospace, chemicals,
engineering, metals, mining and energy/utilities firms. Large companies dominate in
these branches, with the necessary resources to invest in e-procurement and
e-marketplaces. They can adopt e-commerce with both suppliers and consumers.
Marketing/information web site ICT services accounted for Euro 4.45 billion, 39 per
cent of the total European e-business services market in 2000. Web site hosting took
79 per cent, web site design 13 per cent and other ongoing support services 8 per
cent. Web site hosting and design are outsourced in two-thirds or more of the cases. The
role of the government as party to business transactions involves it in electronic commerce
practices as well. Besides the demand/supply for ICT services to support so-called on-
line government, e-procurement adds a strong boost to the uptake of electronic
commerce; public procurement is large relative to GDP in many European countries.
To sum up, the service sector — especially finance and insurance — and large
manufacturing organisations lead in e-commerce applications. Small and medium-
sized companies lag significantly behind, despite high interest in advancing their access
to e-marketplaces. The barriers to entry, however, depend on the costs of adopting
e-commerce applications and on the control of the knowledge to be traded along
digital channels.

41
From E-Commerce to Digital Economy

Optimal Allocation vs. Dynamic Accumulation

Goods, services and coded information are the media for embodying, reproducing
and transferring knowledge from one context to another. Markets, hierarchies and
networks are the three basic knowledge-transfer mechanisms, co-ordinating the
relationship between producers and users of knowledge (Grandinetti and Rullani, 1996).
Transnational corporations were early users of information and telecommunications
applications well before electronic commerce entered the mass media jargon and even
before the adoption of the EDI standard. The wider the networks and the deeper the
relationships between enterprises, the greater the need to ensure replication and smooth
transfer of knowledge within and between firms. This calls for efficient, all-inclusive
and wide-reaching information systems, shared languages and common semantics.
Digital circuits, network technologies and HTML provide the best available means to
that end.
Study of the economic impact of ICTs focuses on the effects of communications
technologies on transaction costs and of electronic brokerage on co-ordination and
search costs as they stimulate electronic integration within the firm and the supply
chain (Malone et al., 1987). That is consistent with a view of the information circulated
and shared over electronic channels as a static component, to be appropriately packed
and relocated within the electronic hierarchy or market in accordance with the principle
of the optimal allocation of resources. Because of the philosophy of decentralisation
and interactivity on which they are based, however, the internet and internet-related
applications enhance the dynamism of the knowledge transfer and creation processes.
By blurring the borders between producers and consumers, goods and services and
markets and hierarchies, the internet produces feedback effects: resistance, adaptation,
creativity and exit.

Network Instability

The effect of internet technologies integrated with ICTs does not fall exclusively
on the reduction of costs associated with business management. It also touches on the
reconfiguration of networks. This is not new. What is new is that ICT and the internet
make knowledge accessible to individuals and organisations as soon as they interact
(i.e. in real time), not exclusively after they have interacted. The market — as a set of
transacting but not interacting individuals — cannot represent fully the dynamics of
knowledge creation and re-creation, which have a fundamental relational nature
independent of the information and communication platform but heavily influenced by it.
A deeper investigation of the implications of B2B electronic commerce for
business models would include business process re-engineering in the value chain,
complemented with two new models — of the value network and of dynamic market

42
configuration — to introduce the concept of market process engineering
(Timmers, 1999). Value networks, focused on deepening the internal relationships
within enterprises by integrating information flows, and dynamic markets, focused
on increasing value from flexible external relationships, have existed for a long time.
Now, however, digital technologies affect how the value-creation process is organised,
entry to and exit from networks and the evolution from one to another network
configuration. The digital market for goods and services opens networks to companies
or networks that are outsiders, without necessarily incorporating them into stable
relationships. Typically, market dynamism enters into networks via ICT and the internet,
changing their configurations but not their nature as intermediate forms of co-ordination
between hierarchy and market. Access to dynamic networks is a precondition for the
development of new businesses, but it is now assessed increasingly by the ability to
compete in the market.

Access to the Digital Economy

The structural change brought about by ICTs and the internet is giving birth to
the digital economy. By increasing information’s divisibility, transportability and
tradability, and by making possible on-line interaction between producers and consumers
of knowledge, ICT and the internet permit a separation of the supply of technological
information from the accumulation of technological competence. (Antonelli, 1999).
That has an extremely important impact on economic development and growth, which
depends increasingly on the quantity of knowledge incorporated into goods, services,
business processes and organisations, and on the degree of “connectivity” in the socio-
economic system.
The debate about what’s new in the “new” economy and which new models will
replace the old ones remains open, constrained by the difficulties of measuring the
“new”, and of interpreting change using “old” paradigms and parameters such as
productivity. The OECD works on measurement, economists produce research
hypotheses, institutions organise high level conferences and journalists stoke the fire.
Some argue that the new economy is new only to mainstream economists (Antonelli
et al., 2000), because the disequilibrium nature of allocation and production processes
and continuous structural dynamics are not new at all in economics. As the digital
economy makes time and knowledge visible as key economic resources, we no longer
face a problem of rational choice, but rather a process of learning (Pasinetti, based on
a quote in IST, 2000).
The digital economy offers as many opportunities as challenges for development.
Old and new players can either take economic advantage of the application of new
information and communication technologies or be disadvantaged by their mis(sed)-
application. ICTs do facilitate fast communications and circulation of information at
a relatively low cost, but the economic impact depends on how technologies and
communication intertwine with other resources, namely knowledge and financial capital.

43
 Entering the digital economy requires a “gateway”, a term that serves as both a
metaphor for and a literal description of access. Access is needed to get into (or not be
excluded from) networks — the ICT infrastructure, business organisations, supply
chains, clusters of companies and virtual enterprises, the research community or policy
institutions. It means having fast and low-cost connectivity in the free market,
interoperable systems, user-friendly applications, open standards and programmes
addressing pluralistic demand rather than requirements for adaptation to market-leading
technological solutions. The characteristic interactivity of the new ICTs is a key quality
that must be preserved. Global diffusion and accessibility of ICT tools (such as operating
systems and basic applications) and exploitation of the full potential of ICTs are
difficult to couple. A subtle confrontation is underway in the market between top-
down pressures towards simplification — supported by policies for global market
diffusion of integrated architectures and tool kits — and bottom-up pressures for
open and flexible platforms, designed to address and capture complexity.
A dual relationship of co-operation and conflict characterises research and the
markets. Market exploitation of research is indispensable to fuel investment in research,
but the need to meet current demand and lead the market may divert research from
innovative paths to replication and trigger lock-in. These effects generate a process in
which technologies selected by the market are not always the best or most useful.
Technology is not neutral. It participates in the social process of fulfilment and
redefinition of users’ requirements (Malerba et al., 1999). Information technology, in
particular, enjoys an extremely powerful role in shaping the market and social relations
because it is simultaneously process, product and industry (Eischen, 2000). In the
digital economy the contribution of ICT to development depends on access to all
three, which in turn means access to:
— network infrastructure built on ICT products and services;
— knowledge generation, codification and trading; and
— network dynamics.

Access to the ICT market

Both individuals and organisations provide the demand. Whatever the application,
the purpose, the equipment, the services mobilised and the value of an e-transaction,
an end-user somewhere clicks the icon “internet connection”. Over 65 per cent of the
372 453 000 regular internet users in the world in 2000 inhabited a macro-area
corresponding to 14 per cent of the world population, i.e. mostly the OECD countries.
The same holds for internet host distribution (OECD, 2001). North America leads by
far — 168.68 hosts per 1 000 inhabitants, compared with 59.16 in Oceania, 20.22 in
Europe, 2.53 in Central and Southern America, 1.96 in Asia and 0.31 in Africa.
Africa, with 780 million people, has about as many hosts on the internet as a small
Eastern European country, Latvia, with a population of 2.5 million (Jensen, 2000).

44
 A small number of American Internet Backbone Providers (IBPs) control access
to the internet. They provide the gateways to about 90 per cent of total backbone
traffic in the United States, where most internet traffic must transit. IBPs provide
access to information located in the United States. They secure access to the core
routing structure and to all internet addresses in the world (Foros and Kind, 2000).
Such centralisation of control of an inherently decentralised network raises the costs
of connection, particularly when there is empty space between end-users, local telecom
operators and the IBPs, as in Africa. There, the high costs of local and international
calls discourage local ISPs from entering the market and operating cross-border
connections. This generates a large cash outflow for transit fees — estimated at
$400 million per year — that reduces funds available for investment in local
infrastructure (UNCTAD, 2000).
A look at the extremes of internet development and underdevelopment between
the United States and Africa makes one appreciate the role of policy as the indispensable
supporter of market maturation in the digital economy. The United States opened the
internet to the market after it had grown up under the protective wing of the state and
when it was ready to fly in a liberalised telecommunication environment. In Africa,
the PSTN revenue transferred to governments has few available substitutes, making it
difficult for them to contemplate wholesale liberalisation and competition.
Several forces should generate an impetus towards appropriate policies in the
underdeveloped areas of the world. They include the fast development of alternative
connection channels bypassing the fixed phone network and the widely perceived
advantage of broad internet access in the less-connected world, together with the
current concern about the digital divide. Policies can exert deep influence on investors’
decisions. Interesting data (UNCTAD, 2000) show that, despite a strong correlation
between the use of the internet and per capita income levels, income is by no means
the sole determinant of use. Some countries with rather low per capita incomes have
a large number of internet hosts.
Figures showing the penetration of ICT are usually taken as plain evidence of
the digital divide, which is approached predominantly in terms of access to the ICT
infrastructure. That is obviously a necessary precondition for any further exploitation
of the potential of digital technologies, but not a sufficient one. Access to the
infrastructure means having the opportunity, on the demand side, to click and use it,
and, on the supply side, to participate in building networks, products and services.
The share of ICT production in gross world product rose from 8.6 per cent in 1998 to
10.4 per cent in 2000. The ICT share of 2000 GDP was 13.3 per cent in Europe,
8.2 per cent in the United States, and 6.7 per cent in Japan. Studies on ICT investment
in the G7 countries show that in 1990-96 the real growth rate of capital investment
was 23.8 per cent in the United States and ranged from 11 per cent to 18.6 per cent in
the other six countries. ICT may contribute to output and productivity growth not
only in the ICT-producing sectors, whose economic weight is growing, but also in

45
ICT-using sectors. Exploratory investigations of the impact of ICT capital goods on
the acceleration of multi-factor productivity (MFP) in non-ICT-producing industries
are furthest advanced in the United States (Bassanini et al., 2000).
The effects generated by equity investment in dotcom companies also deserve
mention. Morgan Stanley, surveying 1 750 US companies, calculated that despite the
sell-off over the past year (a loss of about $1 trillion), technology companies that
have gone public since 1980 have created more than $2.5 trillion of value. Moreover,
Morgan Stanley calculated the overspending on IT as a result of excess funding raised
by dotcom companies in 2000 at around 6 per cent of total IT spending, “not as high
as many believe” (Financial Times, 16 April 2001).
The past five years have seen a rush to reposition in the global market by both
old-economy and new-economy corporations, trying to occupy the newly generated
(or expected) ICT-driven markets through a wave of mergers, acquisitions and alliances,
supported until 2000 by an extremely favourable stock market performance. The
early movers’ strategic priority has been to ensure competitive if not dominant market
shares. The process has been remarkable in the telecommunications industry, leading
to the re-formation of giants amidst deregulation in the United States and Europe.
US companies dominate the IT market. Of the 50 top IT firms, 36 are based in
the United States, nine in Japan, four in Europe and one in Chinese Taipei
(OECD, 2000). The supply side of the digital economy is well rooted and growing in
the United States, Europe and Japan, based on large corporations extending the range
of business activities from original core competencies to new ICT and internet-related
competencies. They battle to control an increasingly global and innovation-led market.
A trend towards business convergence is also underway, mirroring technology
convergence and reflecting the need to ensure a critical mass to sustain rapid innovation.
Fragmented dotcom businesses — those that do not die off — are being absorbed
within the networks of large corporations. Meanwhile, partnerships among
telecommunications companies, IT service providers and consulting companies
configure dynamic networks as an alternative to all-inclusive corporations.

Access to the Knowledge Market

Possession and control of specialised knowledge (whatever the technological

profile and the industry) is not sufficient to ensure competitive positioning in the
market. To the extent that markets are global, the individual company or network of
companies needs to control increasing amounts of information and knowledge to cope
with global competition, even if hitherto it has been active exclusively in the domestic
or regional market. In the interconnected, open market, businesses operating in the
local market confront both local and international competitors, but access to the
networks where market knowledge flows is not as free as the enhanced technical
accessibility would lead one to expect, either for consumers or producers and traders
of knowledge.

46
 The information and knowledge are incorporated increasingly in intangibles
delivered as final or intermediary products by the service industries (Windrum and
Tomlinson, 1999). The impact of ICTs on knowledge transferability goes beyond the
provision of broad, fast, interconnected channels to a logical structure for knowledge
transfer embedded in network software (Eischen, 2000). Information and knowledge
tradability by means of software codification implies:
— awareness of the information and knowledge in one’s possession;
— control of the codification procedures and tools;
— compatibility through adoption of common interface standards.
Despite the trend towards market-led centralisation in leading networks and
procedures, space remains for innovation and competition by players with valuable
knowledge. Knowledge is the output of an accumulation process whose complexity is
due to a pluralistic set of agents and components interacting over different temporal
and social contexts. Micro-diversity and the interaction of different levels of co-
ordination generate the microeconomic turmoil that underlies the “restless” process of
knowledge accumulation. Economic growth can be seen as the evolutionary outcome
of this process (Metcalfe et al., 2001). Local knowledge — knowledge generated
locally that is tacit or codified under different standards — provides indispensable
inputs to the knowledge market, whatever its organisation. The critical testing point
for the correlation between ICT and development lies at the crossroads between
horizontal and vertical applications and standards. They either transform knowledge
into information traded on networks interfacing the global market or re-aggregate it
from local networks to access the global market.

Development in the Digital Economy

The digital economy is in early days. The potential of digital technologies to

support the knowledge accumulation process remains far from full exploitation. The
interplay between the dynamics of institutional resetting and those of micro-behaviour
is an asynchronous, endless process. The most striking opportunity offered by current
ICT development lies in the affordability or perceived accessibility of the digital
infrastructure to an increasingly large portion of the world population, sharing different
cultural and business models and living standards. The creativity bubbling behind the
scene is certain to find its way toward the accumulation of knowledge along the
digital net, as soon as there is a gateway to enter.

Digital Divide

The term digital divide was apparently coined at the World Economic Forum
Annual Meeting in Davos in early 2000, after which the G8 Summit (July 2000)
endorsed the Global Digital Divide Initiative in the Okinawa Charter on the Global
Information Society. The G8 Summit in Genoa (July 2001) subsequently adopted a

47
Plan of Action to reduce the digital divide. The UN, World Bank and OECD have
devoted resources to understanding the phenomenon and designing appropriate remedial
policies (Analysys, 2000, UNCTAD, 2000 and OECD, 2001). The early vision of
digital divide was mostly oriented to the gaps between individuals, households and
businesses, within and between countries, in access to and usage of the internet
(OECD, 2001). Its measurement involves comparisons of indicators related to ICT
infrastructure (connectivity, geographic dispersion, telephone-internet charges,
correlations of income, education, ethnic variables and business size with ICT
penetration, etc.). A stress on the transversal dimension of the digital divide appears
in comparisons between rich and poor countries and within the developed ones as
well. The work of Dot Force (Digital Opportunity Task Force) leading up to the
Genoa Summit stressed that synergistic policy actions must be put in place, targeted
to improve both the status of connectivity and development of local physical and
human capital (G8 Summit, 2001). The basic foundations of the information society
and the digital economy are the same: access to and usage of network technologies.
Policy plays a key role in overcoming the digital divide wherever it exists. It
involves not only capital investment and policy guidelines and regulations to improve
access to the digital infrastructure, but also pro-active policies exerting a catalytic role
on citizens’ behaviour by moving administrative procedures and public services on-
line. Public administration acts as a key lever to promote ICT and the internet, as both
consumer and supplier of on-line services. Moreover, the use of ICT can even facilitate
the solution of endemic problems in the less developed world, such as public administration
transparency, health care, education, logistics and — particularly important — support
policies for local knowledge development via business and network creation.

Local Development

Research on local clusters of firms emphasises their tight link with learning and
innovation, to the extent that clusters support “firms … to collectively transform ‘low
tech’ industries … into tacit knowledge intensive industries and internalise this competitive
advantage within the cluster” (Mytelka, 2000, p. 27). The delocalisation of activities
from developed to less developed countries driven by low labour costs, local business
integration into the supply chains of foreign-based companies and the establishment
of locally-managed subsidiaries by multinational companies are all potential channels
of knowledge transfer and accumulation in the developing world. Other elements are
necessary, however, to upgrade local activities into local development — elements
such as market opportunities, capital investment, local capabilities and policy support.
The right mix cannot be summarised in a few “golden rules”. Approaches differ
in the ICT industry itself, where outsourcing of software development and ICT services
from US, European and Japanese corporations to some less developed countries is a
well-established practice. India, for example, differs from the Philippines or from
Mexico (Lubeck and Eischen, 2000). India provides interesting evidence of ICT-driven
local development as it migrates up the value chain from being a leading offshore
software development centre to ICT services more lucrative than traditional code

48
programming. The Indian software industry as a whole has grown about 50 per cent
annually over the past decade. Besides its comparatively low costs, India has capitalised
on available technically skilled and English-speaking manpower (it has the second-
largest group of software professionals in the world after the United States) and on
deep Indian roots in Silicon Valley technology and venture capital companies. All this
makes India highly competitive vis-à-vis China, Russia, Eastern Europe and the
Philippines for big corporations outsourcing software development and other services
(Financial Times Dossier India, 2001). It has allowed the successful entry of Indian
brands on the ICT market (four Indian companies are listed on the US Nasdaq and
NYSE). The Andhra Pradesh region has revealed that it is possible to turn to innovation-
led local development the same resources that are assets for competition in the global
market. A farsighted government strategy and the integration of local and global
networks at the highest level support local development. The strategy is not one of
latecomer catch-up. It seeks to establish a local information society around a high-
tech pole competitive on a global scale.
There is broad agreement on the existence of locational patterns in the ICT
sector, based on either of two types. The first involves geographical concentration and
clustering in critical metropolitan or intra-metropolitan areas; it has been observed in
the United States, the United Kingdom and Sweden. The second reflects a tendency
towards dispersal as the impact of ICT reduces the “friction of distance”, especially
within the spatial organisation of large companies. The shift from and to centralisation
and decentralisation is a dynamic process, which depends on the nature of the activity
and particularly the nature of the knowledge at stake. Many researchers observe that
upper-tier activities (command and control of knowledge-intensive activities) tend to
be spatially concentrated in nodal centres, while lower-tier activities (based on easily
codifiable and replicable knowledge) may be delocalised to seize lower-cost
opportunities in remote nodes of networks. The impact of delocalised ICT services on
local development in host areas is not well understood.
Essentially, the correlation between digital inclusion and local development
depends on the quality of access to the ICT infrastructure and of the knowledge to be
traded over the digital infrastructure. The dynamics of this correlation may evolve
over time as local activities upgrade technologically. That depends, however, not only
on local human capital and other resource endowments but also on the mix of policy
commitment and strategy able to direct the opportunities offered by the digital
technologies towards social and economic value creation.

Conclusions

The ICT-driven changes currently underway in the economy and society raise
questions for analysts, policy makers and the general public about the impact of ICTs
on development. The questions stem from the perception that the innovations reshaping
communications, business management, consumer preferences and social behaviour
are radical, and their dynamics, still unfolding, can destabilise existing market structures

49
and organisations. These changes have a dominantly but not exclusively technological
nature, deriving from the pervasiveness of ICTs and highlighting the roles of
information and knowledge creation and sharing in most economic processes.
The impact of digitisation on the global communication architecture and on
knowledge transfer makes it sensible to introduce the notion of a digital economy.
The extraordinary growth of supply and demand for goods and services related to the
physical infrastructure represents a major propellant to growth in the countries where
the ICT industry is based. The ICT market is global, but the industry remains based in
the developed world, with some exceptions. The dominant software firms,
telecommunications carriers, ICT equipment makers, e-commerce businesses and
multimedia applications firms at the core of recent US economic growth are mostly in
the United States, Europe and Japan. Nevertheless, the trend towards technology
convergence, the shortage of skilled human resources in the OECD area and the need
to address multicultural global demand give room to local businesses as either suppliers
or brokers of ICT services all over the world.
The potential of ICT to alter business process organisation and network dynamics
is very high and far from fully exploited. The most visible present impacts appear in
adjustment oriented towards cost reduction in the service sector (mostly finance) and
in big supply chains associated with heavy procurement expenditures, where large
manufacturing corporations usually lead. Harder to measure, but nevertheless highly
relevant, is the impact of new ICT in the information-based and knowledge-based
activities that support virtual communities or virtual enterprises in multi-sector value
chains. The worldwide diffusion of the service economy (including in the less developed
world) makes the potential impact of ICT that much more extensive.
Barriers to ICT networks, exclusion from digital marketplaces or communities
and interdiction from the generation and sharing of knowledge and innovation risk
producing irreversible effects — market fragmentation, loss of resources, reduction
of pluralism and slower technical progress and innovation — in both developed and
developing countries. The digital divide is multiform. The link between ICTs and
development depends on access to dynamic networks, to information and knowledge
creation, codification and trading, and to financial capital.
However global the knowledge market, the sources of knowledge have an
essentially local dimension aside from any global input. The correlation between digital
inclusion and local development depends on the quality of access to the ICT
infrastructure and of the knowledge to be traded over the digital infrastructure.
Policy institutions play a key role in piloting the transition to the information
society and the digital economy. Both government and corporate strategies must address
the new challenges as opportunities for development. Besides dealing with access to and
governance of the digital infrastructure, policies can exert a deep influence on investors’
decisions and play a significant role in supporting the local development of network and
knowledge resources needed for effective competition in the global marketplace.

50
 Notes

1. The literature review includes empirical studies carried out by the authors themselves
and others from 1997 to 2000.
2. The centrality of time and preferences has been emphasised by the so-called attention
economy. See Aigrain (1997).
3. “A virtual organisation is a goal-oriented enterprise (i.e. unit or function within a
company) operating under metamanagement ”, (Mowshowitz, 1999, p. 9).

51
 Bibliography

AIGRAIN, P. (1997), “Attention, Media, Value and Economics”, First Monday, 9 December,
available at www.firstmonday.org.
ANALYSYS (2000), The Network Revolution and the Developing World, Final Report for
World Bank and InfoDev, [http://www.infodev.org/library/working.htm].
ANTONELLI, C. (1999), The Microdynamics of Technological Change, Routledge, London.
ANTONELLI, G., N. DE LISO and R. LEONCINI (2000), Innovation and Human Capital in the
New Real Economy: Some Evidence and Interpretative Hints, paper presented at the
40th Congress of the European Regional Science Association, Barcelona, 29 August-
1 September.
BASSANINI, A., S. SCARPETTA and I. VISCO (2000), “Knowledge, Technology and Economic
Growth: Recent Evidence from OECD Countries”, OECD Economics Department
working papers 259, OECD, Paris.
E ISCHEN, K. (2000), “Information Technology: History, Practice and Implications for
Development”, CGIRS Papers Series 2000-4, Center for Global, International and
Regional Studies, University of California, Santa Cruz.
EITO/EEIG-EUROPEAN INFORMATION TECHNOLOGY OBSERVATORY/EUROPEAN ECONOMIC INTEREST
GROUPING, (2001), Frankfurt am Main.
EUROPEAN COMMISSION (EC) (1997), Communication to the European Parliament, the Council
of Ministers, the Economic and Social Committee and the Committee of the Regions,
COM (97) 157final, 15/04/97, at http://www.ispo.cec.be/Ecommerce.
Financial Times Dossier India, July 2001.
FOROS, Ø. and H.J. KIND (2000), “National and Global Regulation of the Market for Internet
Connectivity”, paper presented at the Third Berlin Internet Economics Workshop,
26-27 May.
G8 SUMMIT (2001), “Preparatory work, Report from Session 6 — Digital Divide and
International Co-operation”, 12 March.
GRANDINETTI, R. and E. RULLANI (1996), Impresa transnazionale ed economia globale, La
Nuova Italia Scientifica, Roma.

52
JENSEN, M. (2000), African Internet Status, at www3.sn.apc.org/africa, accessed 19 December,
2000.
KALAKOTA, R. and A.B.WHINSTON (1997), Electronic Commerce, Addison-Wesley, Reading,
MA.
LEFEBVRE, L.A. and E. LEFEBVRE (2000), “E-commerce and Virtual Enterprises: Issues and
Challenges for Transition Economies”, paper prepared for the United Nations
Economic Commission conference on Global Electronic Commerce, Geneva, 19-20
June.
LUBECK, P. and K. EISCHEN (2000), “Silicon Islands and Silicon ‘Valles’: Informational
Networks and Regional Development in an Era of Globalization”, in N. KLAHN, P.
CASTILLO, A. ALVAREZ and F. MANCHÓN (eds.), Las Nuevas Fronteras del Siglo XX1/New
Frontiers of the 21st Century, La Jornada Ediciones, México, D.F.
MALERBA, F., R. NELSON, L. ORSENIGO and S. WINTER (1999), “‘History-Friendly’ Models of
Industry Evolution: The Computer Industry”, Industrial and Corporate Change,
Vol. 8, No. 1.
MALONE , T.W, J. YATES and R.I. BENJAMIN (1987), “Electronic Markets and Electronic
Hierarchies”, Communications of the ACM, Vol. 30, No. 6.
M ETCALFE , J.S., M.D. FONSECA and R. RAMLOGAN (2001), “Innovation, Growth and
Competition: Evolving Complexity or Complex Evolution”, CRIC Discussion
Paper 41, Center for Research on Innovation and Competition, University of
Manchester.
M OWSHOWITZ , A. (1999), “The Switching Principle in Virtual Organisation”, in
Organisational Virtualness and Electronic Commerce, Proceedings of the 2nd
International VoNet Workshop, 23-24 September, Institute of Information Systems,
Department of Information Management, University of Bern, in www.virtual-
organization.net.
MYTELKA, L. (2000), “Local Clusters, Innovation Systems and Sustained Competitiveness”,
I NTECH Discussion Paper 2005, Institute for New Technologies, United Nations
University.
OECD (2001), Understanding the Digital Divide, background documentation for the
OECD Emerging Market Economy Forum on Electronic Commerce, Dubai, UAE,
16-17 January.
OECD (2000), Information Technology Outlook 2000, OECD, Paris.
OECD (1998), A Borderless World: Realising the Potential of Global Electronic Commerce,
OECD, Paris.
PASINETTI, L., based on a quote in IST (2000), The New Economy of the Global Information
Society - Implication for Growth, Work and Employment, Report of a workshop,
Brussels, 6-7 April.
QUAH, D.T. (1999), “The Weightless Economy in Economic Development”, Centre for
Economic Performance Discussion Paper 417, London School of Economics.

53
SCHILLER, D. (2000), Digital Capitalism — Networking the Global Market System, The
MIT Press, Cambridge, MA.
SMITH, M.D., J. BAILEY and E. BRYNJOLFSSON (1999), “Understanding Digital Markets: Review
and Assessment”, at http://www.ecommerce.mit.edu/papers/ude.
TIMMERS, P. (1999), Electronic Commerce: Strategies and Models for Business-To-Business
Trading, John Wiley & Sons, New York, NY.
UNCTAD (2000), Building Confidence - Electronic Commerce and Development, Geneva.
WINDRUM, P. and M. TOMLINSON (1999), “Knowledge-Intensive Service and International
Competitiveness: A Four Country Comparison”, Technology Analysis & Strategic
Management, Vol. 11, No. 3.

54
 Chapter 3

E-Commerce for Development:

Between Scylla and Charybdis
David O’Connor

Introduction

What do the internet and e-commerce mean for people living in developing
countries? For the vast majority of rural people eking out an existence from the land,
the answer is, “probably very little, at least in the near term”. Certainly, for them,
research into how to improve the productivity of their land or how to grow new crops
that offer better nutrition or higher market returns matters much more. This does not
mean that information and communications technologies (ICTs) are irrelevant to poverty
reduction, but rather that the benefits they offer are often rather indirect and long-
term. Arguably one of the most important contributions of ICTs has been to the
productivity of the research process itself. From medical and pharmaceutical research
to crop research, it is difficult to imagine returning to a world without powerful
computers and rapid communications among researchers spread across the globe.
Admittedly, few of the innovations facilitated by ICTs have brought tremendous benefit
to the poor, but even a few major breakthroughs — e.g. low-cost malaria or AIDS
vaccines — could yield enormous benefits.
The focus here lies not so much on these broad developmental applications of
ICTs as on the technologies’ potential contribution to business development and
entrepreneurship in developing countries. In other words, the concern is with
e-commerce1 and what benefits it may offer developing countries, notably as suppliers
of goods and services to the world market.

Digital Opportunities?

While much media attention has focused on the “digital divide” separating rich
from poor countries, the costs of ICTs have fallen at historically unprecedented rates.
Thus, while large income gaps between the richest and the poorest countries still pose

55
major problems of ICT affordability, the prospects have never been so good for poor
countries to acquire advanced technologies relatively early in their diffusion phase.
Rather than widening the technology gap, rapid technological innovation at the frontier
can actually hasten diffusion by inexorably driving down costs of computing power
and communications. Technological innovation is, of course, only part of the explanation
for the rapidly declining costs of telecommunications and internet use. Widespread
deregulation of the telecoms sector in both developed and developing countries has
also played a crucial role.
If the technology is becoming more affordable to developing country
entrepreneurs, is it becoming more useful? The answer to this is a qualified “yes”.
The qualifications are several. First, for many entrepreneurs in developing countries,
information needs are highly localised. Unless content providers have placed useful
local information — on market prices, financing, weather conditions, pest prevention/
eradication, veterinary health, etc. — on the internet, its usefulness is apt to be limited.
Not all locally useful content needs to be provided locally (e.g. global satellite images
and meteorological information downloaded from remote web sites can be useful to
local fishermen or farmers). In localities where few people are educated and even
fewer are literate in English, the usefulness of the internet will likely be very limited.
Even here, however, the internet can be combined with traditional communications
media such as AM radio to disseminate downloaded information over the airwaves.
Other ICTs, notably mobile telephony, can also be valuable in this context, as the
Grameen Phone experience in Bangladesh demonstrates.

D-evelopment E-conomics

Analysis of the economics of the internet in a developing country context ought

to start from the recognition of the role of information in an economy, particularly in
the functioning of markets. The literature on transactions costs (from Coase onwards)
identifies information asymmetries as one important source. Likewise, Stiglitz and
others have made plain how pervasive information incompleteness is, notably in
insurance and financial markets. These imperfections in information flows and access
have given rise to search costs, and they can result in a variety of compensatory
mechanisms and alternative governance structures, e.g. the internalisation of
transactions within a single firm when the costs of transacting through markets become
prohibitive.
Information technologies raise the possibility of smoothing many of the
information flows in any economy. They can make markets work more efficiently,
and they can improve information management within organisations, whether firms,
governments or others. Empirical work on the size of such cost savings is just in early
days, but it is a potentially fruitful field of quantitative research, as internet-based
transactions can generate rich databases at low cost.

56
 A review of early empirical work on internet markets paints a mixed picture of
their relative efficiency vis-à-vis conventional markets (Smith et al., 1999). Four
measures of efficiency are found in the literature: price levels, price elasticity, menu
costs and price dispersion. The most consistent set of results relates to menu costs,
which are found to be lower in internet markets, as reflected in the frequency and the
size of price changes. With respect to price levels, the earlier studies found them to be
higher in internet markets, while more recent ones found them to be lower. A possible
explanation is the change in price behaviour with increasing entry into and maturity
of these markets. On price elasticity, Goolsbee (2000) finds on-line buyers to be
highly price-sensitive, as reflected in the significantly higher use of the internet in US
states with high sales taxes. On the other hand, Lynch and Ariely (1998) find that
providing better product information mutes price competition in an electronic wine
market, making possible a better fit between buyers’ tastes and the available supply.
Similarly, some evidence on price dispersion suggests a significant degree of it in
internet markets. A variety of explanations includes market immaturity, heterogeneity
of on-line retailer characteristics (e.g. trustworthiness, brand name recognition), retailer
market segmentation strategies and price discrimination. In short, precious little
evidence yet shows that internet markets are consistently more efficient than
conventional ones.

Information in Developing Economies

One factor differentiating developing from developed economies is that more

serious information imperfections characterise their markets. Another, related to the
first, is that transaction costs are higher. If this is a fair description of reality, then a
reasonable working hypothesis is that ICTs can potentially yield larger cost economies
in developing countries than in developed ones. This depends in part on the source of
the transaction costs. It is possible that, in many cases, high transaction costs reflect
fundamental institutional weaknesses for which ICTs cannot easily compensate. For
example, the lack of codified regulations, procedures and protocols governing
transactions and/or weak incentives for their enforcement can raise transaction costs.
Each transaction takes on the character of a one-time event, with a low level of
predictability and a possibly high opportunity cost in time spent to complete it. Of
course, excessive but predictable regulations can also impose significant opportunity
costs, but without an additional high probability of failure to complete the transaction.
Assuming that some incentive exists (in the form of competitive market pressures)
to minimise transaction costs, then the availability of ICT to assist in cost reduction
may itself provide a spur to the codification and documentation of information so as
to make possible the technology’s effective application. Without the incentive structure
in place, ICT does not offer a quick technological fix. The importance of competitive
pressures suggests that ICT adoption in developing countries may occur earliest in
those sectors most heavily exposed to international competition — although strong
domestic competition can have the same effect.

57
 Depending on how broadly one defines transaction costs, some may be more
amenable to reduction through the application of ICTs than others. For example, if
the costs result largely from poor inventory management techniques, they may be
more easily reduced through applying ICTs than if they result from poor transport
infrastructure. In general though, the inventory cost savings potential in developing
countries would appear to be large. Guasch and Kogan (2001) find that the costs of
raw materials inventories in manufacturing are two to five times higher in developing
countries than in the United States (see also Blond, 2001, for an India-US comparison
of inventory-to-sales ratios in several manufacturing sectors).
The objection may be raised that, even if markets and other institutions manage
information less efficiently in developing countries, there is less of it involved in most
transactions, so poor information management matters less to overall economic
efficiency. It is intuitively plausible that the information intensity of a largely agricultural
economy is lower than that of a heavily service-oriented economy. Yet, to the author’s
knowledge, there is no good empirical measure of the information intensity of different
markets or economic sectors, independently of a measure of ICT investment or use
per se. Within given sectors, information-handling requirements may actually be greater
in developing than in developed countries. Cement is an example. The Mexican cement
maker, Cemex, faces stronger incentives than its major richer country competitors to
computerise logistics and shift sales and distribution on-line, because it deals with
many small distributors — a common pattern in developing countries — rather than a
few large ones as is more typical in developed countries. (See “E-strategy brief”, The
Economist, 16 June 2001, pp. 85-86). If a less concentrated distribution sector is a
general distinguishing feature of developing countries, then industrial enterprises not
just in cement but across a broad spectrum of industries may face similarly strong
incentives to automate logistics and sales.
One of the touted benefits of the internet and web-mediated e-commerce is the
elimination of intermediate transaction layers between producer and ultimate customer
for a particular good or service (“digital disintermediation”). Yet early evidence on
internet markets (e.g. for used cars) in OECD countries suggests that they may not
eliminate middlemen but merely alter their role and their way of doing business.
Even with ICT, in some kinds of markets there may be reason for continued reliance
on the services of middlemen. For instance, looking at even a 3D virtual model of an
automobile on a computer screen is a poor substitute for test driving it.
The middleman is a pervasive response to information and other market
imperfections in developing countries. What the internet may well do is put more
information — on consumer tastes, prices of competing products, etc. — in the hands
of small producers in these countries, so that their bargaining power vis-à-vis the
middleman is enhanced. If there is continued need for such a market intermediary
— e.g. to handle transport and logistics or financing — then at least s/he no longer
stands to profit from imperfect information. Goldstein and O’Connor (2000) present
some examples of this disintermediation in markets for handicrafts, but in several
cases non-profit organisations acting as intermediaries ensure that a larger share of
revenues stays with the producers. Intermediaries are not eliminated altogether.

58
 Information in an Evolutionary Perspective

Industry structure and supply chain organisation are crucial determinants of

information management requirements, hence of the likely benefits to be derived
from ICT application. Structure and organisation are dynamic, however, so the
information needs of a given industry and enterprise are likely to evolve over time.
Industries tend to be rather concentrated when they are young, become less concentrated
with new entrants eager to share in the high profits, then become more concentrated
again in maturity, as economies of scale favour size and the more profitable either
obliterate or absorb their competitors. Similarly, the degree of vertical integration not
only varies systematically across sectors (in accordance for example with differences
in transaction costs) but also can vary over time. When an industry is still relatively
immature, it may not yet have a group of specialist supplier firms, so existing firms
need to provide their own internal capacity — e.g. for component, tool and/or
machinery making, which eventually can be spun off or contracted out to independent
firms.
Does vertical dis-integration of the supply chain itself give rise to greater demand
for ICTs and electronic networks? In all probability it does, since one common feature
of dis-integration is the emergence of multiple independent suppliers of any system or
component, increasing the number of potential information flows. Here again one
might ask whether there is a systematic tendency for industries in developing countries
to be more or less vertically integrated than their developed-country counterparts.

Path-Dependent ICT Evolution

Industrial technology, industry structure and their historical evolution shape the
way in which ICT is used and how important it is to an industry’s costs and
competitiveness. Two contrasting examples illustrate the point: automobiles and
personal computers (as epitomised by the “Dell direct model”) (see Helper and
MacDuffie, 2000). To begin with computers, several technical features condition the
global organisation of production and the place of ICT and the internet in organising
the supply and distribution chain. These include:
— discrete modules coterminous with sub-systems;
— standard interfaces;
— easy outsourcing of module production; and
— easy integration of modules in final assembly.
This implies that a personal computer can be customised at relatively low cost
simply by re-specifying the combination of standard modules incorporated in the
final product. If the customer prefers a more powerful processor or hard disk drive,
or a CD-ROM instead of floppy disk drive, they can be readily substituted one for
another, each fitting into the same standard slot.

59
 In contrast, the auto industry model is built on the following technical
considerations:
— distributed systems (e.g. the power train);
— only a few stand-alone modules (like the “cockpit”);
— multiple components, many model-specific; and
— integral design.
This requires close co-ordination between the big assemblers (or OEMs
— Original Equipment Manufacturers) and their major (tier-1) suppliers, who are
often involved in module and/or system design. On the other hand, it implies less
flexibility than with personal computers to configure the final product according to
customer preferences. The “bespoke” vehicle is still too expensive for the mass market.
Even within the automobile industry, contrasting models of supplier relations
may condition how auto companies make use of the internet. The models depend in
part on national industry characteristics, in part on enterprise culture and in part on
location within the vertical supply chain. In the spirit of Hirschman, Helper and
MacDuffie (2000) refer to these as the “exit model” and the “voice model”. The exit
model is characterised by relatively low switching costs between suppliers, leaving
them with weak bargaining power vis-à-vis a given OEM customer. The forward
auction, in which suppliers bid against one another to offer the most attractive price,
is associated with this model. The voice model is characterised by intensive information
exchange between supplier and customer, with a correspondingly higher cost of
switching suppliers. This gets manifested in long-term collaboration on joint vehicle
design. In the exit model, the internet becomes the site for conducting parts auctions;
in the voice model, it becomes the medium for intensive data interchange among
collaborators.
Once more, the question arises of where developing-country entrepreneurs fit
into the broader scheme of things. Between the personal computer industry and the
auto industry, there would seem to be greater opportunities for outsourcing in the
former, also greater opportunity for new entrants to produce standardised components
for world markets. This is the case with semiconductors, disk drives, terminals and
other PC components and peripherals. The quality requirements for participating in
such markets are daunting, however, especially as suppliers to computer industry
leaders. In the auto industry, national and regional market requirements are sufficiently
varied (in terms of vehicle size, fuel-efficiency requirements, road conditions, etc.)
that maintaining a “local” presence can be important. While a few large developing
countries have strong domestic OEM and 1st tier supplier bases, in most countries
these industry segments are heavily dominated by large multinationals. Local suppliers
are mostly confined to tiers two and below, i.e. precisely those segments of the industry
where the “exit model” can be most readily applied. For them, the arrival of the
internet and on-line auctions poses the risk of diminished bargaining strength vis-à-vis
their customers and diminished profit margins.

60
 B2B2C

The union of B2B and B2C e-commerce is one of the most promising sources of
efficiency gains the internet can offer. By rationalising inventory control and supply-
chain management, the internet should permit better demand forecasting all along the
chain, from final customer back to raw materials. In principle, this effect can be
measured by reference to the variation in inventory-to-sales ratios of firms along a
vertical supply chain.
E-commerce can, in theory, place greater power in the hands of consumers to
customise products and services according to their own preferences. The evidence on
internet prices vs. conventional market prices suggests that, rather than being solely or
even principally a vehicle for price arbitrage, the internet is often used as a means of
better matching supply to demand. The “customer as designer” model may not be
equally well suited to all products and industries, but it is a tendency in many. The
feedback from the customer interface should be valuable all along the supply chain,
not only in inventory management as mentioned above but also in adapting to and
anticipating changes in consumer preferences, incorporating new design features,
planning more effective advertising and marketing campaigns, etc.
One of the principal challenges facing developing country export producers is
keeping abreast of major OECD market trends. This is especially relevant for
differentiated product markets such as textiles and clothing and footwear, but also for
those in which product innovation is very rapid, such as semiconductors, PCs and
other electronics. Having timely information on changing market conditions and having
the R&D and/or design capacity to respond in a timely fashion to those changes are
clearly not the same thing. Still, presence in a given market presupposes a certain
technical capacity, one that the internet can help to enrich and update. Where the
internet can be especially useful, however, is in providing the rapid feedback from
customers visiting a B2C web site.

Alternative Internet Pricing Models

There are two basic approaches to pricing of goods and services on the internet.
The first is fixed pricing, the on-line equivalent of the supermarket checkout counter,
which is why it is appropriate for Amazon.com to provide a virtual shopping cart to
its on-line customers. The other is dynamic pricing, with real-time price setting either
through negotiation or an auction. The principal attraction of fixed pricing is for
small, individual, low-value transactions in which economies of scale on the supplier
side permit volume discounts on the buyer side. There may also be some cost shifting
from the seller to the buyer, who must bear any search costs as well as the costs of
providing information needed to complete the transaction. Presumably the purchaser’s
willingness to absorb these costs is bounded by the alternative cost of buying the same

61
product in a “bricks-and-mortar” shop. The main attraction of dynamic pricing is for
high-value transactions where negotiation costs are a small share. The high market
liquidity in this case also permits a closer matching of demand and supply.
Instances of both pricing models can be found in developing countries, although
casual empiricism suggests that the dynamic pricing model is far more pervasive
across a broader range of markets than in OECD countries. The village market
epitomises this; few prices are fixed and almost any purchase is the object of haggling
— although clearly both buyer and seller have reservation prices. If the internet is
indeed to become the marketplace of a “global village”, the pricing model even for
standardised products may need to move in the direction of a dynamic one. Moreover,
in developing countries tight firm and household budget constraints normally mean
that purchases are made in small quantities on an as-needed basis, rather than in Costco
quantities to capture volume discounts. Given this fact, developing countries may
well pioneer micro-payment schemes for internet micro-purchases.

Terra Incognita

Even if we should succeed in navigating between the Scylla of techno-pessimism

and the Charybdis of unfounded optimism, researching the internet and e-commerce
in developing countries takes us to a vast terra incognita waiting to be explored. The
widespread use of the internet for e-commerce is a very recent phenomenon even in
the OECD area, with only a few years of historical data on which to base any quantitative
analysis. Given the low internet penetration rates and the low e-commerce turnover in
most of the developing world, quantitative analyses of the type performed thus far
mostly in the United States still lie some way in the future.
The internet is a global technology creating a global e-commerce marketplace.
Developing countries have differing capacities to respond to the challenges and
opportunities it presents. Public policy plays a role in the response, but mostly in
providing a suitable regulatory framework to encourage private investment, both in
the provision of the basic telecommunications infrastructure and in the operation of
internet-related businesses such as web portals and internet service providers. The
private sector has taken the initiative to deal in a virtual environment with many of the
problems encountered by conventional businesses (for example consumer protection),
plus new ones thrown up by the internet itself (e.g. security of transactions). Still,
supportive government legislation and regulation (e.g. on litigation of disputes,
recognition of electronic signatures, personal privacy) are needed to underpin private
initiatives. Co-operation is also required among countries to address jurisdictional
questions and potential conflicts in such areas as enforcement of contracts, tax liability
and privacy protection. Given its longstanding involvement in crafting guidelines on
various policy aspects of e-commerce for its Members, the OECD as an organisation
is in a favoured position to offer guidance to developing country governments on
issues of e-commerce readiness.

62
 In parallel with policy guidance, empirical research is needed on the ways in
which ICTs, including but not limited to the internet, affect the competitive environment
in sectors where developing countries have traditionally enjoyed a comparative
advantage. Garment manufacturers in Bangladesh, coffee growers and traders in Kenya,
auto components suppliers in Brazil — all need to know what the internet and
e-commerce will mean for them. Similarly, potential e-ntrepreneurs in developing
countries would benefit from knowing how the internet is changing various service
markets and what opportunities this may open for outsourcing to developing countries.

Notes

1. The term e-commerce is used here to refer to all applications of electronic network
technology (notably the internet) to a firm’s business operations, whether to deal
with customers, suppliers and other partners, or to organise internal management
functions (the latter application is sometimes referred to in the literature as
e-business).

63
 Bibliography

BLOND, C. (2001), “Analysis of the Determinants of Transaction Costs”, OECD Development

Centre, Paris, mimeo.
BRYNJOLFSSON, E. and B. KAHIN (Eds.), (1999), Understanding the Digital Economy, MIT
Press, Cambridge, MA.
GOLDSTEIN, A. and D. O’CONNOR (2000), E-Commerce for Development: Prospects and Policy
Issues, Development Centre Technical Paper No. 164, OECD, Paris.
G OOLSBEE , A. (2000), “In a World Without Borders: The Impact of Taxes on Internet
Commerce”, Quarterly Journal of Economics, Vol. 115, No. 2.
G UASCH, J.L. and J. K OGAN (2001), “Inventories in Developing Countries: Levels and
Determinants — A Red Flag for Competitiveness and Growth”, Working Paper 2552,
World Bank, Washington, D.C.
HELPER, S. and J.P. MACDUFFIE (2000), “E-volving the Auto Industry: E-Commerce Effects
on Consumer and Supplier Relationships”, prepared for E-Business and the Changing
Terms of Competition: A View from Within the Sectors, Roundtable sponsored by the
Berkeley Roundtable on the International Economy and Haas School of Business,
U.C. Berkeley, 12 September version.
LYNCH, J.G., JR. and D. ARIELY (1998), “Interactive Home Shopping: Effects of Search Cost
for Price and Quality Information On Consumer Price Sensitivity, Satisfaction with
Merchandise, and Retention”, at Marketing Science and the Internet, INFORM
College on Marketing Mini-Conference, Cambridge, MA, 6-8 March.
SMITH, M.D., J. BAILEY and E. BRYNJOLFSSON (1999), “Understanding Digital Markets: Review
and Assessment”, at http://www.ecommerce.mit.edu/papers/ude.

64
 PART TWO

SECTORAL ANALYSES

65
 Chapter 4

The Prospects and Challenges of E-Business for the

South African Automotive Components Sector:
Preliminary Findings from Two Benchmarking Clubs
Sagren Moodley

Introduction

The South African automotive components industry currently finds itself under
threat from three directions. First, rapid and sweeping liberalisation of the national
trade policy regime1 has increased international competition. Second, the domestic
market has become stagnant. Third, local sourcing by South African original equipment
manufacturers (OEMs) is eroding rapidly as they become increasingly integrated into
the global strategic operations of their parent companies. According to Barnes (2000b,
p. 70), “(U)ncompetitive firms with poor international linkages will disappear from
the (components) industry, but those firms that improve their competitiveness and
create appropriate linkages with international firms could benefit from burgeoning
export sales”. The industry thus urgently needs to reposition itself in global-scale
value chains in order to consolidate relationships with OEMs and facilitate exports.
Transnational companies’ (TNCs) control of international production and trade
networks makes it extremely difficult for domestic component firms to export
independently. Hence, the issue of connection to OEMs or to TNC 1st tier component
suppliers has become fundamentally important for long-term survival and growth.
From a development perspective, export growth prospects for the components
industry hinge increasingly on leveraging information and communication technologies
(ICTs) to promote industrial upgrading within global-scale value chains. In the global
automotive industry, TNCs are moving quickly to integrate global supply chains
electronically. Most are setting up procurement and supply chain systems. The South
African components industry will need urgently a capability to use the internet to take
part in these global production networks. In a few years’ time, its firms may find
themselves bidding for business within a global, internet-connected automotive industry.

67
 The industry’s need to enter export markets and compete with new entrants in
the domestic market has become critical to its long-term survival and growth.
Information flow is one critical mechanism through which firms could improve or
consolidate their positions within the value chain. Gaining access to global-scale trade
networks is crucial, and digitally connecting to advanced markets has critical importance,
whether through complex sourcing relationships or the export of finished products.
The use of the internet to co-ordinate production through domestic and cross-border,
inter-firm networks will likely have a significant impact on the industry’s competitiveness.
The internet’s potential to create seamless, collaborative supply networks has recently
come under academic scrutiny (Gereffi, 2000). Network-oriented ICTs, by compressing
space, time and knowledge, allow for unprecedented reach, speed and complexity in the
management of the automotive supply chain. Theoretically, the internet holds great
potential for revamping traditional supply chains to improve data flow and streamline
operations. The challenge for South African component firms is to position themselves
within the evolving internet-connected global supply chains in order to gain competitive
advantage and capture value. Drawing on evidence from 19 component firms in the
Eastern Cape (ECBC) and KwaZulu-Natal (KNBC) Benchmarking Clubs, this paper
explores the uptake, potential and challenges of e-business for the sector.
The paper aims to provide a preliminary analytical foundation to help focus the
policy debate. Section two, a synopsis of the key challenges facing the industry, stresses
the importance for survival of an outward orientation, global connectedness to
transnational production and trade networks, the generation of export volumes and
industrial upgrading. Section three introduces the two benchmarking clubs, the research
population for the study. It provides an overview of the clubs and flags some of the
members’ supply-chain problems, which give scope for e-business to streamline the
supply chain and reduce information-related uncertainties. Section four presents a
conceptual definition of e-business and discusses its importance for the sector. Section
five speculates on the internet’s likely impact on the components value chain, especially
for supply-chain management. E-business is still in an early stage of development and
the current research base is quite thin. With the theoretical foundation for the empirical
research thus established, section six reviews the empirical findings from a survey of
the 19 firms in the two benchmarking clubs. Section seven concludes the study.

The South African Automotive Components Sector: An Overview

Except for the German-owned OEMs and their global lead-source component
suppliers, the South African automotive industry generally has weak global networking
links. The inward orientation of its components sector reflects a history of state
protectionism and import-substitution industrialisation (ISI) during the apartheid era
(Joffe et al., 1995). Trade isolation, disinvestment and economic sanctions during the
1980s and early 1990s reinforced this inward focus. Nationally based producers were
long insulated from the cut and thrust of international competition. Since the transition
period (post-1994), however, the twin pressures of globalisation and rapid trade-policy

68
liberalisation have substantially altered the industry’s landscape. In September 1995,
the government launched a Motor Industry Development Programme (MIDP) to
promote greater integration of the domestic automotive assembly and component
industries into the global automotive market. It aimed primarily to improve the
international competitiveness of the sector and to expand it, especially through exports.
The industry involves multiple players in logistically complex supply chains. It
can best be described as a “producer-driven” supply chain with multi-layered production
systems organised hierarchically into tiers (i.e. OEMs and 1st, 2nd and 3rd tier component
suppliers) (Gereffi, 1999). Figure 4.1 clearly shows that except for a few independent
aftermarket suppliers, seven OEMs largely control the industry, namely Toyota SA,
Volkswagen SA, BMW SA, Mercedes Benz SA, Samcor, Automakers and Delta. The
domestic aftermarket currently provides enough scope and volume to sustain a number
of independent component manufacturers of relatively stable-technology products such
as air filters and batteries. Relationships within and between the automotive value chains
tend to be fixed, linear and clearly demarcated. Enormous potential exists, therefore,
to create an environment of more direct, cost efficient and interactive relationships.

Figure 4.1. South African Automotive Value Chains: A Schematic Overview

OEM Controlled

Potential impact of
E-Business on the
automotive value chains Original Aftermarket
OEM market Equipment for automotive
Seamless electronic
inter- and intra-firm for passenger Supply (OES) parts & accessory
communication and & commercial market for sales through
information flows vehicle sales replacement parts independent
& accessory sales retailers &
ICT architecture based through the OEMs
on open networks repair shops
underlined by partnerships
and trust in the value chain
Electronic B2B transactions
High degree of ICT
Infrastructure integration 1st Tier 1st Tier 1st Tier
and interoperability Component Component Component
Supplier Supplier Supplier
High capacity ICT
networking capabilities
Digital buyer-supplier
linkages 2nd Tier 2nd Tier 2nd Tier
Component Component Component
Internet-enabled supply Supplier Supplier Supplier
chain management and
logistics
ICT-enabled global
connectivity 3rd Tier 3rd Tier 3rd Tier
Component Component Component
ICT-enabled value chain Supplier
upgrading Supplier Supplier

ICT-enabled innovation and

learning curves

69
 The South African automotive industry is the 18th largest in the world but
accounts for less than 1 per cent of the world market. It employs about 82 000 people
and accounts for 6.4 per cent of manufacturing GDP (Barnes, 2000b). About
180 component firms primarily supply the industry, with another 200 secondary
suppliers (Barnes and Kaplinsky, 2000a). The components sector is being forced to
adjust to fundamental changes taking place in the assembly industry. Locally owned
and controlled domestic assemblers are re-incorporating under global ownership, and
the OEMs are integrating increasingly into their parent companies’ global manufacturing
operations (Barnes and Kaplinsky, 2000a)2. Toyota SA and the Delta Motor Corporation,
the two exceptions, still have majority local ownership and are still largely inwardly
focused3.
Vehicle and component manufacturers are becoming increasingly dependent on
exports. In 1999, vehicles, parts and accessories comprised 6.5 per cent of total exports;
this category did not even feature in 1990 (Finance Week, 27 October 2000). The
liberalised trading environment has effectively set the domestic OEMs free from local
purchasing requirements, and they now operate under a more or less duty-free tariff
structure. The net effect has been a recent trend for domestic OEMs to import
components previously sourced domestically (Barnes and Kaplinsky, 2000b). Survival
of the components industry will depend to a large extent on finding a space in
increasingly international production networks, largely controlled by TNCs. Thus,
greater connectivity of domestic firms to TNC component firms, which effectively
control design, marketing and technology for high value-added products, is critical
(Barnes and Kaplinsky, 2000a,b; Barnes, 2000a,b).

The Benchmarking Clubs

The Eastern Cape (ECBC) and KwaZulu-Natal (KZBC) Benchmarking Clubs

have as their prime objective the continuous improvement of their members’ operational
competitiveness through the generation of comparative domestic and international
benchmarks. They are effectively joint partnerships between the Department of Trade
and Industry through the Sector Partnership Fund and local automotive component
firms. Club members seek to improve their competitiveness through a learning network.
Figure 4.2 provides an indication of how important the component manufacturers are
as a source of employment in both KZN and the Eastern Cape. It shows the average
employment figures for 13 of the 19 club members4. Average employment peaked in
1996 (at 455 for the Eastern Cape and 263.5 for KZN), then steadily declined to lows
of 216.04 in KZN and 335.0 for the Eastern Cape in 1999. Figure 4.3 shows average
turnover figures and hints at the potential contribution that the clubs make to the
regional and national economies5. Average ECBC turnover peaked in 1997 (at
R 99 737 000), and deteriorated steadily thereafter. By contrast, average KZNBC
turnover improved continuously from 1996 (R 47 345 922) to 1999 (R 60 779 499).

70
 Figure 4.2. Average Number of Employees
(Component Manufacturers)

500

400
No. of Employees

300

200

100

0
1994 1995 1996 1997 1998 1999

Eastern Cape (N=5) KZN (N=8)

Source: IRP Database.

Figure 4.3. Average Turnover

(Component Manufacturers)
120

100
Rand (millions)

80

60

40

20

0
1995 1996 1997 1998 1999
Eastern Cape (N=3) KZN (N=8)
Source: IRP Database.

Inventory-level measurements provide a rough proxy for the level of integration

in the supply chain. Figures 4.4-4.7 compare the inventory performance of firms
belonging to the two clubs with their international counterparts, to present an outline
of the variance in inventory performance. The comparison is based on close
benchmarking of like-for-like component manufacturers. The key inventory control
measures considered are total inventory (Figure 4.4), raw materials inventory
(Figure 4.5), work in progress (Figure 4.6) and finished goods inventory (Figure 4.7).

71
 Figure 4.4 clearly shows that while improvements in the number of days of
total inventory have been significant, the general performance of club members remains
well below that required for international competitiveness. Moreover, while raw
materials (Figure 4.5) and work in progress (Figure 4.6) inventories have on average
improved significantly, finished-goods holdings (Figure 4.7) have significantly
deteriorated at many firms. In most cases, however, this reflects increased export
growth rather than actual inventory deterioration; many club members need increased
finished-goods stock to meet monthly or biweekly shipping schedules to foreign
customers. Nonetheless, the figures clearly show that club members have a supply-
chain management (SCM) problem, in that they lag behind their international
counterparts in inventory performance6. Inventory, excess capacity and stock-outs are
the key consequences of information-related uncertainties in the value chain. E-business,
in theory, offers club members the possibility of closer synchronisation of the supply
chain. They can reduce inventory and shorten cycle times as firms jointly manage
inventories across the supply chain, forecast sales together and plan collaboratively to
keep levels of component and completed product inventory down to a bare minimum.

Figure 4.4. Number of Days of Total Inventory:

Club Members vs. International Counterparts
80

Club members (n=12)

70
66.77 66.08
60
58.88 61.29
50 46.84 52
40.02 40.90
Days

39.01
40 35.05
International firms (n=12)
30

20

10

0
1995 1996 1997 1998 1999
Source: IRP Benchmarking Data.

72
 Figure 4.5. Raw Material Inventories:
Club Members vs. International Counterparts

45 Club members (n=12)

40
39.44 39.59
35
34.49
30 31.16
25.61
25 23.65
Days

22.32 26.94
20.76
20 18.07
International firms (n=12)
15
10

0
1995 1996 1997 1998 1999
Source: IRP Benchmarking Data.

Figure 4.6. Work in Progress: Club Members vs. International Counterparts

16

14 Club members (n=12)

14.63 14.51
12
12.23
11.27
10
Days

8 7.40 8.93
6.63 6.88
6.20
6 5.34
International firms (n=12)

0
1995 1996 1997 1998 1999
Source: IRP Benchmarking Data.

73
 Figure 4.7. Finished Goods Inventory:
Club Members vs. International Counterparts

20
18 Club members (n=12)
18.87
16
16.14
14 12.81 12.82 14.27
Days

11.70 12.05 11.64

12
12.16
10 International firms (n=12)
9.77
8
6
4
2
0
1995 1996 1997 1998 1999

Source: IRP Benchmarking Data.

E-Business: Towards a Definition

Pervasive ICT use, deregulation, the opening of markets and global trade
expansion drive economic globalisation (Cohen et al., 2000; Dicken, 1998). ICTs
now form an integral part of its accelerated pace, linking nations into complex webs
of transnational exchanges (Castells, 1996; Gereffi, 2000)7. The internet is becoming
a key enabler of the global, digitally networked economy, with economic progress
becoming increasingly knowledge-driven, and information and knowledge becoming
primary wealth-creating assets (Castells, 1996; Evans and Wurster, 2000; Fine, 1998).
Although the internet’s precursor appeared in the late 1960s, e-business is primarily a
product of six significant transformations in the global economy:
— the globalisation of markets;
— the shift towards an economy based on knowledge and information;
— the growing prominence of ICTs in the economy;
— innovations in business organisation and practice (just-in-time production, total
quality management, knowledge management, etc.);
— the liberalisation of the telecommunications sector, primarily in OECD countries;
— technological innovations such as email, the world wide web, internet browsers
and the expansion in the volume and capacity of communication networks (viz.
optic fibre, digital subscriber line technologies and satellites).
The term “e-business” has no widely accepted definition. In a very broad sense,
it means doing business over the internet. This paper defines e-business as any form of
commercial or administrative transaction or information exchange that takes place

74
via an internet-based, computer-mediated network. E-business thus entails the
application of the internet to the complete value chain of business processes. It places
a premium on openness, transparency and trust. The internet offers a wide spectrum
of potential commercial activities and information exchange (Figure 4.8). The paper
focuses exclusively on business-to-business (B2B) internet interactions (the shaded
block in Figure 4.8), for two main reasons. First, the concern is primarily with the
potential of the internet for enhancing inter-firm supply-chain management, and,
second, because current trends indicate that B2B e-commerce will far outstrip business-
to-consumer (B2C) e-commerce (Intelligence, “Business in the Internet Age”,
May 2000). One reason for the exponential growth anticipated in B2B e-commerce in
the automotive industry is the expected migration of supply-chain management from
relatively expensive, proprietary (i.e. closed) electronic data interchange (EDI) networks
to the open network communication system of the internet.

Figure 4.8. E-Business Matrix

Government Business Consumer Employee

Government G2G G2B G2C G2E

Business B2G B2B B2C B2E

Consumer C2G C2B C2C C2E

Employee E2G E2B E2C E2E

B2B e-commerce refers to procurement, logistics and administrative processes

occurring between firms. Companies use the internet to integrate the value-added
chain, which can extend from the supplier of raw materials to the final consumer.
Inter-business e-commerce can be divided into two categories: open, marketplace-
based trade and direct trade between business partners. The former occurs in various
internet-based auctions or exchange sites (e.g. Covisint), while the latter occurs either
through a firm’s web site with an on-line purchasing function or an EDI-type network.
Some argue that B2B e-commerce will likely spread globally and grow rapidly
primarily because of its potential for: i) controlling business costs (associated with
inventories, sales, procurement and distribution); ii) connecting to markets through
greater geographical reach; iii) value creation; iv) productivity and systemic efficiency
gains in the value chain; and v) advanced supply-chain management (Cohen et al., 2000;
IBM, 2000). It appears, therefore, that the South African automotive components
industry could, in theory, benefit substantially from adopting e-business.

75
Potential Impact of E-Business on the Automotive Component Value Chain:
Theoretical Evidence

In the simple, physical value chain, the flow of goods is linear. The traditional
value chain is a series of arms-length, value-adding links from raw materials to end
customers. It consists of a sequence of activities with clearly demarcated points of
input and output. This value-chain model (Figure 4.9) deals with information as a
support to the value-adding process and not as a source of value itself. Contrast this
with the internet-enabled value chain (Figure 4.10) where the value-adding steps are
performed through and with information. The internet-enabled virtual value chain is
non-linear (Figures 4.10 and 4.11). Rayport and Sviokla (1994, 1995) argue that
firms compete in both the physical world of goods (the marketplace) and the virtual
world of information (the marketspace). The virtual value chain consists of “a matrix
of potential inputs and outputs that can be accessed and distributed through a wide
variety of channels” (Rayport and Sviokla, 1995, p. 83). Information can be captured
in real time at all stages of the chain, analysed electronically to improve performance
at each stage and co-ordinated across the chain. In a completely networked value
chain, the internet connects every node (i.e. the “pull” system becomes operational)
(Figure 4.10). Materials are pulled through the supply chain ultimately by customer
request. The logic implies that firms should hold minimum inventory and make just
what is needed when it is asked for and not before.

Figure 4.9. A Simple Physical Value Chain

Design Production Outbound Consumption

Raw Materials, Logistics, &
Inbound Logistics, Marketing Recycling
Transforming Inputs, &
Packaging, Sales
etc.

Physical Value-Adding Stages

Linear and Serial rather than Interactive

76
 Figure 4.10. A Virtual Corporate Value Chain

Web-Enabled Enterprise
Resource Planning (ERP)

Web-Based ICT Architecture

Digital Convergence of ICTs

Supply Side Demand Side

Research & Development Outbound Logistics
Design Marketing
Intranet
Financials & Human Resources Sales
Workflow
Operations Planning & Execution
Purchasing
Inbound Logistics Innovation
&
Raw Materials Creativity
Production Processes The Corporate Portal

Government Suppliers Customers Markets Subcontracting 3rd Party

Networks Agencies
Characteristics of the Intra-Organisational Information System:
A corporate portal: a centralised home page with links to various departments and services.
Includes corporate procedures, rules and regulations; as well as corporate news, views and vision.

The home page is used as the entrance to a network designed to pull all the old “legacy” computer
systems together.

Seamless information flows through tightly integrated ICT links, which promote connectivity
and collaboration within the enterprise (B2E), including remote employees.

ICT-enabled internal and external processes to improve efficiency and effectiveness

within the company.

Streamlined internal communications and business processes. The intranet affects the way
managers communicate with staff, and staff with each other.

An information-integrated channel: a high level of overlap and inter-dependency

between front-office and back-office systems.

The intranet provides a way to build a corporate culture to bind employees together.

Changes in the organisation of work toward flatter hierarchies, the decentralisation of

decision making and semi-autonomous project-based teams. The intranet enhances team
building and encourages (within the company) the emergence of horizontal communities
bound together by a common function or interest.

The intranet is a quick way to help employees come to terms with changing business
methods and terms of competition.

Knowledge management: involves efficiently connecting employees who know with employees
who need to know and converting personal knowledge into organisational knowledge. The
intranet provides the ideal platform for pooling the creativity, skills and knowledge of a workforce.

Running online training programmes for employees.

Focus is on getting information to key decision makers.

77
 Figure 4.11. An Integrated Value Chain

Overlap indicates flexible, responsive digital links between

the value-adding stages

Raw Materials

End Consumer

Web-based IT integration of the upstream and downstream businesses

High-Capacity Networking Capabilities

Gereffi (2000, p. 2) argues that “The new digital era of globalization is

characterized by an explosion in connectivity that is melting the informational glue
that holds corporate value chains and global value chains together”. With virtual
integration, the links no longer stand alone. Instead, strategic, information-defined
inter-relationships proliferate between the various players in the value chain
(Figure 4.12). Cross-enterprise virtual integration blurs the traditional boundaries and
roles of the players in the value chain. The linkages between firms tend to become
more fluid, interactive, pervasive, customised, spontaneous and potentially mutually
beneficial. Further, the digital links enable the firm to become more flexible and
responsive to change. The virtual value chain is an integrated system (Figure 4.11)
rather than “a set of discrete though related activities” (Figure 4.9) (Rayport and
Sviokla, 1995, p. 78). Firms use the internet to integrate the value-added chain, which
can extend from the supplier of raw materials to the final consumer (Figure 4.11).
Figure 4.11 illustrates how an integrated internet-based system enables a firm to
visualise its value chain from end to end, as an integrated whole. The firm can add
real value by focusing on what it does best and then collaborating with other firms
involved in the overall supply chain. The internet facilitates collaboration both within
and between firms by dynamically sharing information in real time (Figure 4.12).
Figure 4.11 also shows how the internet enables the networked or virtual organisation
to connect, dynamically and in real time, the supply and demand sides of the company.
Companies use intranets to link their employees, providing them with a new
communication platform and a new way to access up-to-date information relevant to
them. The next step would be an extranet, to give suppliers and customers access to
real-time and relevant information by linking into an enterprise’s internet system.
This makes it easier for the company and its constellation of suppliers, customers and
partners to work together more effectively (Figure 4.12).

78
 Figure 4.12. E-Business-Enabled Extended Enterprise

Seamless, Internet-Enabled Information Flows

Virtual Links: Web-Based Interface

Business-to-Business (B2B) The Enterprise Business-to-Customer (B2C)

Supply Chain Links Communication Links

Integrated supply chain Deployment of an intranetb E-commerce (i.e. e-sales)

management which streamlines the firm's
internal business functions, Customer service
E-commerce and connects individual
(i.e. e-procurement) workstations (including Customer relationship
remote employees) management
Stock management
Web-based internal business Interactive marketing
Integrated forecasting process efficiency and
effectiveness Demand forecasting
Quality systems management
Web-enabled enterprise Orders management
Extension of a firm's intranet resource planning (ERP)
to suppliers (i.e. deployment Extension of a firm's
of an extraneta) Internet-connected front- intranet to customers
and back-office systems (i.e. deployment of an
Move from independence and processes extranet)
to interdependence
Cross-functional orientation Etc.
Etc.
Use of collaboration and
knowledge management tools

A move to responsive
knowledge workers

Etc.

Inter-Firm
Collaboration

Suppliers High Speed Networking Capabilities Customers

A Powerful Virtual Extended Enterprise (Extranet)

Value Generating Stream

Notes:
a) Internet-based network for use by a company and its business partners.
b) Internet-based network for company use only.

79
 The firm focuses on its distinctive capabilities, then forms strategic alliances
with others in the value chain. Increased dependence upon suppliers thus becomes a
requirement and has a major impact on the buyer-supplier relationship. In the global
automotive industry, this relationship, traditionally arm’s-length and adversarial, is
now moving towards closer collaboration as a defence against the complexity and
uncertainty of markets (Humphrey, 1999).
The internet holds great promise for the South African automotive components
sector in three key areas: i) increasing the efficiency of internal processes (Figure 4.10);
ii) streamlining inter-firm linkages (systemic efficiencies); and iii) connecting to markets
(Figure 4.12). Dell Computer’s internet-enabled build-to-order business model, for
example, illustrates the power of a fully integrated internet-based assembler production
system (www.dell.com; Magretta, 1998). Apart from improving intra-firm and inter-
firm process efficiencies, the internet can also play a key role in facilitating supply
chain learning and innovation. B2B internet-based collaborative interaction and real-
time communication are likely to sharpen the competitive edge of the participating
firms, reduce information asymmetries and improve the quality of information
embodied in business relationships.
There are also risks associated with e-business, i.e. cost cutting, price-based
supplier relationships and competitive switching. Electronic links theoretically make
it easier for buyers to begin and end supplier relationships. Component suppliers will
lose the protection afforded by familiar inter-firm relationships and their customers’
incomplete information about alternative supply sources. The pursuit of short-term
price advantages and the concomitant “factory hopping” and “competitive switching”
on the part of assemblers and 1st tier component manufacturers is, of course, at odds
with upgrading the positioning of South African component manufacturers in local
and global value chains. Therefore, building trust, loyalty and reciprocity to reduce
the threat of opportunism and make it difficult to break long-term network relationships
is important.

Empirical Evidence

Several key findings emerged from the survey of 19 South African component
manufacturers. Table 4.1 shows that all club members have access to the Internet, but
only 63.2 per cent have web sites. Connection with the internet, however, does not
mean that its use is extensive. Nonetheless, club members claimed to be using it for a
wide variety of purposes. They are only just beginning to do so for supply-chain
management. Current use is most extensive for intra-firm and inter-firm communication
and for marketing, rather than for inter-business transactions. Respondents cited
inadequate ICT systems integration and a lack of standards for sharing B2B e-commerce
data as the main barriers to fully-fledged e-business.

80
 Table 4.1. Uptake of E-Business
(19 respondents)

E- Business Technology:
Internet 100.0 per cent
EDI 73.7 per cent
Website 63.2 per cent
The internet is used mainly for:
Communication Sales
Order Processing Marketing
Production Procurement
Engineering/Technical applications Services and support
Inventory management Management of distribution channels
Logistics Research

The researchers conducted an internet survey of club members’ web sites, the
results of which show that they are invariably not much more than front ends, on-line
catalogues with orders being e-mailed, faxed and/or taken over the telephone.
Customers cannot use them to check, for example, on service call status, order shipping
status and delivery information. Four firms use their sites to take orders and deliver
products but have not added capabilities such as customisation or interactivity to
distinguish the service from other types of direct selling. This does not really constitute
e-commerce because the firms have not developed fully transactional internet sales
channels. Apparently, too much emphasis goes on establishing a web presence and too
little on ensuring that the ICT infrastructure can support on-line procurement, trading,
marketing and sales. Most club members are at the stage of basic on-line “brochureware”.
The web sites have static information, with minimal ability for interaction beyond
e-mail, company background and, in a few cases, placing orders.
A substantial 73.7 per cent of club members have an EDI system in place. EDI
entails considerable investment in dedicated software and proprietary networks. For
effective inter-firm communications, buyers and suppliers need matching software
and hardware, a common language for communications, a standard layout for documents
and a common communication network. A major limitation is the one-to-one nature
of EDI transactions. Therefore EDI is not suitable for companies engaged in occasional
(or ad hoc) purchases, or even regular low-volume business — exactly the kind of
business that SMEs do. EDI’s major drawback is that it is uneconomical to link small
suppliers into the system. Its cost and supply-chain management advantages create a
tendency for large companies to purchase from suppliers with existing, compatible
EDI links.
These forces tend to narrow and concentrate supply chains within a quasi-co-
operative framework, which could, in principle, limit the scope of a firm’s sourcing
options. Nevertheless, companies with experience using EDI will probably make the
transition to internet-based e-business to streamline the supply chain faster than
companies with no such experience. In this sense, B2B e-commerce via the internet is

81
an extension of the existing EDI network. In contrast to the private interactions of the
EDI structure, the new virtual, open marketplace will likely increase competition
among suppliers and push down purchasing prices.
Box 4.1 indicates the key drivers of e-business in the South African automotive
component sector, as identified by the club members themselves. Box 4.2 lists what
club members perceive to be the major benefits of the internet. Club members apparently
are, by and large, quite positive about the need for e-business, especially regarding
the strong forces promoting its rapid uptake (Box 4.1) and the expected benefits of
internet use (Box 4.2). Table 4.2 reveals that internet-enabled supply-chain management
is very much in its infancy in the benchmarking clubs. This sluggishness reflects the
caution with which enterprises view an ICT tool that is still developing. Many club
members feel that a misplaced step into the B2B space will result in more than the loss
of substantial investment; they fear that they are also “betting the firm” on the move.

Box 4.1. Key E-Business Drivers in the ECBC & KZNBC

á Globalisation
á Trade liberalisation
á Customer requirements
á Closer integration with customers & suppliers
á Increased competitiveness
á Connectivity
á Agility/business flexibility
á Joint ventures
á Market expansion
á Customer retention
á Improved efficiencies
á Cost control
á Customer & supplier information exchange
á Accelerated speed to market
á The need to streamline business processes
á Spread of the internet

Box 4.2. Benefits of the Internet According to the ECBC & KZNBC

á Reduces the cost of information

á Improves inventory management
á Greater geographical reach: increases domestic and
international market access
á Gives the firm a competitive edge
á Reduces transaction costs
á Shortens order-ship-bill cycles
á Speed combined with flexibility

82
 Table 4.2. Internet-Enabled Supply Chain Management
(19 Respondents)
Percentage of Respondents
Replying “Yes”
Use of the internet for SCM 31.6
Do your suppliers have access to real-time information on your company’s sales 10.5
and stock levels?
Do your customers have access to real-time information on your company’s stock 31.6
levels?
Are your firm’s internal operations electronically integrated with those of your 21.1
customers?
Are your firm’s internal operations electronically linked with those of your 15.8
suppliers?
Does your company have the e-business capacity to access your suppliers’
Production capacity? 21.1
Available inventory? 21.1
Lead times? 21.1
Delivery flexibility? 21.1
Are your front-office and back-office systems electronically integrated? 31.6
Does your company require suppliers to make use of e-business technologies? 26.3
Does your company have a supply-chain development programme? 0.0
Does your company use e-business technology for B2B procurement? 78.9
Does your company use e-business technology for B2B sales? 47.4

Integrating functions across businesses, such as workflow arrangements in a

supply chain, is predicated on working, integrated internal systems. Table 4.2 reveals
that many component manufacturers do not have fully automated and integrated back-
office and front-office systems. The sector faces a major challenge to converge legacy
ICT systems with the internet. The supply-chain efficiencies promised by the internet
depend on full internet-based integration of the value chain, to offer a broad range of
collaboration and integration between trading partners. Reaching this ideal will be
expensive and require industry-wide co-operation not generally prevailing in the
industry today. Component suppliers embarking on fully internet-enabled systems at
this early stage in the development of e-business might find the return on investment
quite low. The benefits might not justify the costs in the short term. Yet as more
suppliers and customers use the internet to web-enable their front-office and back-
office systems (internal functionality — Figure 4.10), and to connect with their trading
partners (collaboration functionality — Figure 4.12), the benefits of B2B e-commerce
will likely become more pronounced. E-business must be placed in a relative context.
How fast is adoption compared to competitors? A B2B uptake rate lower than that of
competitors may result in declining value added and market share. Until sufficient
numbers of their main domestic and foreign customers and suppliers participate in e-
business initiatives, individual firms have little incentive to develop a web-based
customer and supplier interface and engage in e-business themselves.

83
 Unlike the TNC 1st tier suppliers, domestic ones do not take responsibility for
upgrading and co-ordinating their own supply chains. In his survey of 46 South African
component suppliers, Barnes (1998) found very little supply-chain development activity
from 1st tier suppliers down to the 2nd and 3rd tiers. Moreover, Barnes and Kaplinsky
(2000a) describe the supply-chain upgrading activities of the South African OEMs as
“modest”. This means that possibilities for upgrading the ICT capabilities of component
suppliers, especially lower-tier firms, remain very limited. This is likely to constrain
the widespread diffusion of e-business in the components sector.
The issue of on-line B2B trading exchanges has recently received a great deal of
attention in the automotive industry; especially with the launch of Covisint, the giant
Detroit-based global virtual marketplace for procuring car components, supplies, services
and information8. Only 36.8 per cent of club members are currently part of an on-line
B2B trading hub (Table 4.3), but note that actual trade via the hubs has thus far been
very limited. Nonetheless, club members listed a number of benefits likely to emerge
from these networks. They also recognised a number of possible risks and stressed
three main conditions for a trading hub to be successful: openness and transparency,
trust and loyalty, and a critical mass of value-chain participants.

Table 4.3. Internet-Based B2B Trading Hubs

(19 respondents)

⇒ Is Your company linked to an Internet-based B2B trading exchange?

Yes 36.8 per cent
No 63.2 per cent
⇒ What are the benefits of the B2B trading exchange?
á Set standards
á Cut costs
á Speed: Internet-based transactions are instantaneous, interactive and rich
á Supply chain efficiency: enables direct interaction between manufacturers and suppliers,
eliminating the need for intermediaries. This is likely to reduce procurement costs
á Market transparency: by being brought into one open platform, suppliers will be able to bid for
contracts they may not previously have known about
á Improves operating efficiencies
á Economies of scale: the system should enable buyers with similar requirements to link
together to order large quantities of inputs jointly, creating economies in production that will
be reflected in lower purchasing prices
⇒ What are the disadvantages of the B2B trading exchange?
á Rationalisation of the supplier base
á Potential oligopolistic control and manipulation by transnational assemblers and TNC 1st tier
component suppliers
á Privacy concerns
á Pressure on prices
á Threat to established partnerships, relationships & alliances
⇒ What are the main conditions for a successful B2B trading hub?
á Openness & transparency
á Trust & loyalty
á A critical mass of participating firms

84
 Box 4.3. Barriers to the Adoption of E-Business (ECBC & KZNBC)

á Low use of e-business by customers & suppliers

á More pressing corporate priorities
á Security concerns
á Lack of standards
á Senior executives who do not support e-business
á Uncertainty over the returns which e-business will deliver
á Lack of e-business skills
á A lack of clear understanding of the potential of the Internet on
the part of top management

E-business is not yet a strategic imperative in the South African automotive

components sector. Most firms in the industry cannot support e-business ventures
because they do not have the integrated customer and supplier ICT interfaces in place
(Table 4.2). Box 4.3 lists the barriers to the adoption of e-business that club members
have identified. The author believes that the primary impediments are:
— the arm’s-length and adversarial nature of current relationships between value-
chain participants, both vertically and horizontally;
— the overwhelmingly cost-driven rather than knowledge-driven nature of the
automotive value chain;
— a pervasive “wait and see” mindset among club members; and
— evolutionary path dependencies that have locked firms into an insular, inwardly
oriented way of thinking. This manifests itself in firms’ reluctance to bring
their customers, suppliers and business partners inside the “corporate machine”.
What is needed therefore is a new strategic direction in the industry, which is
likely to entail a sea change in business mindsets. The progression of component
firms to higher levels of e-business capabilities will likely be slow and incremental.
Two critical success factors in e-business centre on the networking of internal and
external business interactions. Component suppliers will need to seize and rapidly
exploit them in order to be globally competitive. First, strengthening the bonds between
an enterprise and its value chain through an internet-based system is important. Second,
firms need to focus on their intra-organisational connectivity in order to tie transactional
flows together within the organisation. Currently, most of the component firms still
focus primarily on a functional orientation (Level 1 in Table 4.4) or on operational
efficiency in the enterprise (Level 2). None of the firms uses e-business to increase
the organisation’s effectiveness outside the enterprise by linking across the internet
with suppliers and customers to create virtual supply chains (Level 3).

85
 Table 4.4. Levels of E-Business Integration

Levels 1 2 3
Sphere Functional orientation Integrating across Cross-enterprise involvement
functional departments
Rationale Departmental focus, Integrated business A virtual ecosystem that connects
for example: EDI — activities via employees, suppliers and customers by
purchasing & sales internet/intranet extending existing EDI. The extranet aims
department; individual applications. to: build trust and increase customers
departments satisfaction; increase collaboration and
developing specific knowledge sharing between customers and
Internet applications, suppliers; and maximise synergies to lower
e.g. a marketing costs, improve efficiencies and increase
website. quality.
Levers Technological Business processes Cultivating knowledge workers; developing
infrastructure and (process efficiencies and exploiting intellectual capital to create
software applications within the firm) opportunities; and building relationships

Conclusion

The critical mass needed to reap positive network externalities is not yet present
in the South African automotive components sector. It requires leadership by several
of the powerful OEMs and the large 1st tier suppliers to define e-business standards
for all value-chain participants and create incentives that attract more companies. The
catalytic role that early adopters can play is also critical. The following four additional
factors will potentially lead to quicker adoption of e-business in the sector:
— a defensive reaction to competitors engaging in e-business;
— pressure by OEMs and 1st tier suppliers on all of their suppliers to link into
their e-business systems as a condition of trade;
— on-line trading portals achieve a critical mass of members and become established
as trading platforms of choice; and
— the cost-reduction and value creation potentials of the internet begin to be realised
in practice.
The components sector critically needs an outward orientation to generate the
economies of scale and scope required for global competitiveness. Significant and
sustainable growth of the South African automotive industry is contingent upon its
ability to compete in a global operating environment. Moreover, as TNCs integrate
the internet into their cross-border business operations, South African firms run the
risk of exclusion from global value chains if they cannot establish electronic ties with
the major players. The policy challenge for the components sector, therefore, is how
to use the internet to:

86
— leverage, consolidate and deepen its links with the global economy;
— access global internet-based trading portals;
— take advantage of the potentials of globalisation; and
— exploit the systemic and productivity-enhancing possibilities inherent in e-business.
E-business might well become an order-qualifying criterion for producers as a
prior condition for participating in global markets. It could be seen as an advantage
when competing for global supply contracts, and an e-business capability will likely
become influential in determining export success. E-business is still at a very early
stage in its development. The notion that internet application may lead to a sustained
higher level of economic efficiency, still very much at the level of theory, will need
rigorous exploration in practice. Achieving such efficiency gains depends on a number
of factors, including access to e-business systems and the needed skills, firms’ willingness
to open their internal systems to suppliers and customers and, more generally, firms’
evolutionary path dependencies. By not making the transition to internet-based business,
component firms may be placing themselves at risk of becoming less competitive in
the globally interconnected market, impacting on both their current market positions
and long-term viability. If this were to happen, it would have adverse development
implications for the country, such as job losses, a decline in revenue growth and an
eroded export base.
The pressure on automotive component firms to reorganise to tap the advantages
of e-business is currently not high. This is likely to change soon, as e-business becomes
a powerful competitive tool for component suppliers to maintain what Sturgeon (2000)
calls “geographic agility” and “output agility”. The internet holds great potential for
integrating the backward and forward linkages in the automotive components value
chain, and digitally connecting all phases of the value chain from raw material supply
to design, production, marketing, distribution, consumption and finally recycling
(Figures 4.10 and 4.11). Hence, e-business could play an important role in facilitating
network-based, dynamic learning and innovation curves in the value chain and
improving the ability of automotive component firms to move to more profitable,
knowledge-intensive, higher value-added economic niches in international trade
networks. Industrial upgrading could take place through harnessing and exploiting
the information flows and learning potential associated with internet-based buyer-
supplier links. Thus, the internet-connected digital network could, at least in theory,
be used to create a new source of competitive advantage for the sector in export-
oriented development. All these benefits are still only potentials. They must be sought
for and released.
Automotive component firms will have to weigh the importance of protecting
existing relationships, which account for most of their current revenue, against the
advantages of establishing future strategic positions and revenue streams through e-
business. Internet impact can be fully examined only as a long-term phenomenon.

87
Initial assessments, especially at such an early stage of e-business development, are at
best inconclusive. A need exists for long-term analytical studies by independent
researchers.
OEMs developing long-term obligations and investing in their suppliers’ ICT
capabilities are critical for crafting a knowledge-intensive value-chain upgrading path
for automotive component manufacturers in South Africa (Sako, 1992). OEMs need
urgently to adopt a conscious policy of building stable, long-term relationships with
their suppliers based on the transfer of complex information across the inter-firm
link. Stronger inter-firm “informational” and “knowledge-based” relationships between
OEMs and their suppliers are essential for component suppliers upgrading into quality-
driven market segments which offer high returns. The upgrading challenge appears to
be tightly bound with the e-business challenge of creating dynamic, high performance
networking between buyers and suppliers to cope with the “new competition”
(Best, 1990; Porter, 1990).
The ideal situation occurs when a company’s internet-based infrastructure supports
all aspects of the extended electronic enterprise, enabling the firm to inter-operate and
collaborate with any number of business partners (Figure 4.12). In South Africa,
fully internet-enabled supply chain models are a long way off. There is no blueprint
or silver bullet that firms can use to be ready for e-business. They need to go beyond
the traditional view of just looking at their own corporate capabilities and understand
that their growth and profitability will increase by managing and improving the end-
to-end process that delivers products to customers. The internet is not in itself a
panacea or a quick-fix solution for all supply-chain difficulties. The idea is to use the
technology’s reach and ubiquity to enable firms to get closer to their employees,
customers, suppliers and business partners.
E-business capabilities will not, by themselves, automatically lead to greater
domestic and export sales volumes for South African component manufacturers. The
future of the domestic automotive components industry also depends heavily on
achieving international competitiveness through revitalising internal performance
capabilities in conformity with world-class manufacturing principles. Nevertheless, a
failure by domestic component manufacturers to adopt e-business will undermine
their ability to survive the onslaught of global competition and further squeeze them
into the low value-added manufacturing segment of the automotive value chains,
where the barriers to entry are low and price-based competition is most intense.

88
 Notes

1. At a rate much faster than that required by WTO regulations.

2. This is especially true of the German OEMs, i.e. BMW, Mercedes Benz and
Volkswagen.
3. Toyota SA is 72.2 per cent locally owned, and the remaining 27.8 per cent is owned
by the Toyota Motor Corporation of Japan. Delta, on the other hand, is 51 per cent
locally owned, and 49 per cent is owned by General Motors (Barnes and
Kaplinsky, 2000a).
4. Twelve KZN club members and seven Eastern Cape club members.
5. Recall that the data do not include six component manufacturers. Some club members
regard turnover figures as private. Hence, the missing cases.
6. SCM examines how information can be used to change how and when products are
moved to increase efficiency. Companies use SCM to link sales, marketing,
distribution, manufacturing processes, and customers, suppliers and business partners
together for a more unified approach to the market. The establishment of these links
means that firms are able to build relationships with customers, suppliers and carriers
more effectively to reduce operating costs, improve customer service and expand
into new markets.
7. Castells (1996), for instance, writes about a globalised and networked “informational”
capitalism.
8. Covisint has been developed by General Motors, Ford and DaimlerChrysler, and has
recently been joined by Renault-Nissan (www.covisint.com).

89
 Bibliography

B ARNES , J. (2000a), “Changing Lanes: The Political Economy of the South African
Automotive Value Chain”, Development Southern Africa, Vol. 17, No. 3.
BARNES, J. (2000b), “Global Trends in the Automotive Industry: Their Likely Impact on
South African Automotive Assembly and Component Manufacturers”,
Transformation.
BARNES, J. (1998), “Competing in the Global Economy: The Competitiveness of the South
African Automotive Components Industry”, CSDS Research Report 13, School of
Development Studies, University of Natal, Durban.
BARNES, J. and R. KAPLINSKY (2000a), “Globalization and the Death of the Local Firm? The
Automobile Components Sector in South Africa”, Regional Studies, Vol. 34, No. 9.
BARNES, J. and R. KAPLINSKY (2000b), “Globalisation and Trade Policy Reform: Whither the
Automobile Components Sector in South Africa”, Competition & Change, Vol. 4.
BEST, M. (1990), The New Competition, Harvard University Press, Cambridge, MA.
CASTELLS, M. (1996), The Information Age: Economy, Society, and Culture. Vol. I: The Rise
of the Network Society, Blackwell, Oxford, UK.
COHEN, S. J., B. DELONG and J. ZYSMAN (2000), “Tools for Thought: What is New and Important
about the ‘E-Conomy’”, BRIE Working Paper 138, University of California, Berkeley.
D ICKEN, P. (1998), Global Shift: Transforming the World Economy (3rd edition), Paul
Chapman, London.
EVANS, P. and T.S. WURSTER (2000), Blown to Bits: How the New Economics of Information
Transforms Strategy, Harvard Business School Press, Boston, MA.
F INE, C.H. (1998), Clockspeed: Winning Industry Control in the Age of Temporary
Advantage, Perseus Books, Reading, MA.
GEREFFI, G. (1999), “A Commodity Chains Framework for Analysing Global Industries”,
Background Notes for IDS Workshop on the Spreading of the Gains from
Globalisation, Institute of Development Studies, University of Sussex, Brighton,
UK, 15-17 September.

90
GEREFFI, G. (2000), “Beyond the Producer-Driven/Buyer-Driven Dichotomy: An Expanding
Typology of Global Value Chains with Special Reference to the Internet”, unpublished
paper, 14 September, http://www.ids.ac.uk/ids/global/bella.html.
HUMPHREY, J. (1999), “Globalisation and Supply Chain Networks: The Auto Industry in
Brazil and India”, in G. GEREFFI, F. PALPACUER and A. PARISOTTO (eds.), Global Production
and Local Jobs, International Institute for Labour Studies, Geneva.
IBM (2000), Making E-Business Deliver: Business Guide, Caspian, London.
JOFFE, A., D.E. K APLAN, R. KAPLINSKY and D. LEWIS (1995), Improving Manufacturing
Performance: The Report of the ISP, UCT Press, Cape Town.
MAGRETTA, J. (1998), “The Power of Virtual Integration: An Interview With Dell Computer’s
Michael Dell”, Harvard Business Review, March-April.
PORTER, M. (1990), The Competitive Advantage of Nations, Macmillan, London.
RAYPORT, J.F. and J.J. SVIOKLA (1995), “Exploiting the Virtual Value Chain”, Harvard Business
Review, November-December.
RAYPORT, J.F. and J.J. SVIOKLA (1994), “Managing in the Marketspace”, Harvard Business
Review, November-December.
SAKO , M. (1992), Prices, Quality and Trust: Inter-Firm Relations in Britain and Japan,
Cambridge University Press, Cambridge, UK.
STURGEON, T. (2000), “How Do We Define Value Chains and Production Networks?”, IPC
Working Paper 01-005, MIT, Cambridge, MA.
WEB SITES:
www.covisint.com
www.dell.com

91
 Chapter 5

Local Entrepreneurship in the Era of E-business:

Early Evidence from the Indian Automobile Industry
Andrea Goldstein*

Introduction

Few statements are as often heard in management circles as “Supply chains face
fundamental change because of electronic commerce” (Cohen and Agrawal, 2000).
The integration of mass production and mass distribution late in the 19th century
heralded the era of the giant corporation run by professional managers. The upcoming
second supply-chain revolution, structured around web-based exchanges and hubs,
interactive trading and advanced optimisation mechanisms, and matching algorithms
to link customers with suppliers for individual transactions, will have profound
implications for corporate organisation and industry structure. Thanks to open
architecture, supply chains will become trading communities (supply webs) where
rich amounts of information can be shared easily and inexpensively (Box 5.1)1.
Completed on-line transactions are just the tip of the trade iceberg. Less visible
but not necessarily less important are the efficiency gains from using internet technology
as an enabler to optimise common business processes in the whole value chain, including
product development, procurement and supply-chain management. Potential
productivity gains expected from business-to-business (B2B) electronic commerce
include cost efficiencies from automation of transactions, the probable advantages of
new market intermediaries, consolidation of demand and supply through organised
exchanges and changes in the extent of vertical integration of firms (Lucking-Reiley
and Spulber, 2001).
___________________________________
* The author thanks Fiat Auto, Fiat Group, Fiat India, and their executives for kind
assistance and co-operation and Anthony D’Costa, Mahrukh Doctor, John Humphrey and
seminar participants at the International Conference on “E-commerce for Development:
Reviewing Early Experiences, Comparing New Ideas”, the OECD Development Centre
and IE-UFRJ for precious comments and suggestions on an earlier draft.

93
 Box 5.1. Supply Chains and Electronic Commerce
Products reach customers through chains of retailers, distributors, wholesalers,
manufacturers and component suppliers. Supply-chain management is intended to
accelerate the flow of goods, information and capital in both directions along the
chain’s entire length and to help companies monitor that flow. This includes sourcing
and procurement, production scheduling, order processing, inventory management,
transportation, warehousing and customer service. Importantly, it also embodies the
information systems necessary to monitor all of these activities. Successful supply-
chain management co-ordinates and integrates all of these activities into a seamless
process. It embraces and links all of the partners in the chain. In addition to the
departments within the organisation, these partners include vendors, carriers, third-
party companies and information systems providers. The OECD Information, Computer
and Communications Policy (ICCP) Committee defines electronic transactions as
“…the sale or purchase of goods or services, whether between businesses, households,
individuals, governments and other public or private organisations, conducted over
computer-mediated networks. The goods and services are ordered over those networks,
but the payment and the ultimate delivery of the good or service may be conducted on
or off-line” (OECD, 2000). Because the costs of managing the supply chain
— inventory, the warehouse and distribution centre and freight — can represent 10 per
cent to 15 per cent of sales in most industries, the savings that B2B exchanges promise
could have a genuine impact.

This paper documents the use of e-business solutions for supply-chain

management in the automobile industry in India and its consequences for the interactions
between foreign and local companies2. The choice of this sector responds to various
factors. First, the car industry has traditionally pioneered in the incorporation of new
management and organisational techniques, from mass production Taylorism in the
1920s to lean manufacturing and just in time (JIT) in the 1980s. In very recent years
automotive manufacturers have reached a high level of sophistication in outsourcing
the design, manufacture, and assembly of increasingly complex systems
(e.g. Antiblocking Brake Systems, ABS) and modules (e.g. the vehicle cockpit), thereby
shrinking the immediate supply base to sets of so-called Tier I suppliers. Ideally these
assume responsibility for modular assembly and the on-line final module assembly
into the vehicle, take an investment stake in the operation and manage the module
supply chain. While transaction costs may have decreased, the co-ordination of
competencies and knowledge has become more complex. Second, car assemblers are
among the largest multinational companies in terms of assets located outside their
home markets — indeed five of them count among the world’s top 11
(UNCTAD, 2000). Third, and obviously related, producers of motor vehicles
consistently rank among the largest investors in developing countries. In India, since
the liberalisation process began in 1991, the auto sector has received 7.81 per cent of
the total foreign direct investment inflow, the highest share in manufacturing. Fourth,
the industry is an early adopter of e-business and e-commerce. Transport equipment
occupied first place in the 1999 ranking of on-line sales in the United States, with

94
20.8 per cent of shipments originating in e-commerce3. Finally, the technical talent
for setting up e-business systems is probably more widely available in India than in
most other emerging economies.
The challenges for firms in developing countries are daunting, especially as the
increased pace of technological and organisational change is accompanied by the
liberalisation of trade barriers (such as minimum domestic content requirements, limits
on the imports of second-hand cars, non-tariff barriers and investment incentives) that
shielded them for most of the post-war period. On home markets, competition from
imports and from foreign investors located in the developing world has increased very
substantially, while firms selling into external markets are forced to compete with
sophisticated producers to attract the loyalty of increasingly discerning customers4.
The destiny of developing-country enterprises, however, is not cast in stone. In a
business environment dominated by global competition and rapid technological change,
firms show a surprisingly great heterogeneity in their (idiosyncratic) reactions to
changes in the global incentive regime. For automotive components firms, competitive
requirements include strength in core competencies, willingness to invest in new
technologies and R&D, success in quality control and synergistic relationships with
Original Equipment Manufacturers (OEMs). Yet while upgrading to achieve world
standards may be within the reach of individual companies, sustaining this competitive
process requires a systemic effort both downwards and upwards (there is indeed strong
empirical support for the hypothesis that customers’ demand is the strongest factor
pushing firms to improve their performance). Realising efficiency gains from supply
management will not result naturally from routine interaction among firms — not
least because of the large network and co-ordination economies that cannot be
appropriated at the level of the single firm. Lead-firms, usually large and (at least in
developing countries) most often foreign-owned, can play an important role in tutoring
smaller local suppliers (SDS, 2000). Government policies are also crucial in fostering
development of domestic entrepreneurship. The process of opening is showing, perhaps
paradoxically, that the firms better positioned to operate in the new environment are
those that made better use of public intervention in the past (Tewari, 2001).
The paper starts by examining some major trends in the global automotive
industry, then speculates in some detail on the implications of e-business for market
structure and small and medium-sized enterprises (SMEs). It then sketches the evolution
of the Indian car sector in the 1990s, following the progressive removal of most
policy barriers and the entry of foreign OEMs. A case study follows, of Fiat and its
strategy to produce in India the same “world car” that it developed in Brazil and now
sells in other large, emerging markets. How do the speed and timing of business
processes change? How do communications patterns shift — both within the firm and
with external parties (suppliers, customers, alliance partners, etc.)? What are the social
and technological factors that facilitate and constrain transitions to electronic modes
of operating? The limitations of a case study are clear, but they are largely offset by
the richness of the insights from focusing on a single firm, especially when it represents
the larger universe of foreign investors in India. The paper concludes by indicating
implications for policy makers and avenues for further research.

95
Major Trends in the Automotive Industry

After decades of operating a classic push system of manufacturing, building

cars in huge numbers at costly and vertically integrated plants with highly unionised
workforces (co-ordination inside the firm), the world’s leading carmakers have found
themselves saddled with over-capacity and declining profit margins. In the 1990s the
business has become increasingly global, with emerging markets now representing a
quarter of world supply. In some countries (e.g. Indonesia) the indigenous industry
was basically intended to restrict foreign currency outflows, whereas in others
(e.g. Brazil) it eyed participation in the world market from the beginning (Mukherjee
and Sastry, 1996). The divide nowadays seems more a function of the degree to which
countries on the OECD periphery are being incorporated into the productive structures
of MNCs, with Mexico and Eastern Europe on the one hand and China, India and
Mercosur on the other (Humphrey, 1999). A second challenge involves developing costly
new fuel technologies to meet ever-tougher emissions legislation and vehicle recycling
targets while also incorporating new electronic technologies. Another shock, whose
real impact is still unclear, has been the arrival of the internet, a tool that all carmakers
hope to harness in areas ranging from purchasing to in-car equipment to retailing.
OEMs are adapting through three different but closely intertwined strategies5.
The first includes the restructuring of assembler-supplier relations and the adoption
of the modular assembly system. In the old business model characterised by low
levels of trust and high transaction costs, production was internalised and individual
components were awarded on short-term contracts to multiple suppliers selected solely
on the basis of price. Today, Tier I suppliers have become increasingly involved with
customers in the provision — and sometimes design — of complex sub-systems such
as dashboards, rear axles, body panels, seats, etc6. Modularity is preferred because it
requires less investment, reduces time-to-market, allows more flexibility to change
the body type and leads to higher performance in resistance and energy absorption. As
OEMs spread their activities around the globe, suppliers have developed policies of
“follow sourcing”, producing components and complete functions wherever their
customers set up bases. OEMs do indeed try to source production of all components
on an exclusive basis to a reduced number of suppliers, which design and therefore
own the specifications. These organisational linkages find their physical counterpart in
suppliers conducting operations around or inside the car-manufacturing factory. Obviously
the likelihood of trading different components internationally differs, reflecting both
transport-to-value costs and tooling and set-up costs (Barnes and Kaplinsky, 2000).
The second strategy is to reduce the number of vehicle platforms OEMs rely on,
while trying to satisfy consumer demand for a multiplicity of new model variants
— from lifestyle vehicles to sports utilities, from urban two-seaters to people carriers —
which have a shorter life span. A facet of this rationalisation trend in vehicle platforms
is the design of “world cars” adapted to the particular operating environment of
emerging economies, including weather conditions, environmental considerations, road
infrastructure, personal tastes, etc. (Barnes, 1999). This phenomenon too is at play on
a global scale, with identical cars produced around the world (so-called “follow design”).

96
 It is increasingly evident that only the very largest and financially muscular
carmakers — which in many cases are, or have remained until recently, controlled by
founding families (e.g. Ford, Fiat, Peugeot, Volvo, BMW) — will survive on a global
basis. Cash flow considerations are important because the cosy tradition of captive
dealers and exclusive distribution arrangements is doomed, while the margins made
from selling finance and insurance are much higher than from making and selling
cars. Modular assembly also puts a premium on companies that master competency in
a range of related products, a capacity that is achieved faster through joint ventures,
mergers and acquisitions than by organic growth. Hence the third response is a “great
mating dance” of alliances and acquisitions. OEMs, moreover, have spun off their in-
house component activities, first by encouraging them to compete for business and
then by severing ownership links. Increasingly globalised itself, the components industry
is following the same route and, in consolidating, the largest firms take over the most
competitive companies in emerging markets.

The Impact of E-commerce in the Automobile Industry

General Features

One can hardly overestimate the importance of managing the supply chain in
the automotive industry. On average, automobiles contain some 5 000 parts sourced
from some 300-500 companies. The industry is a heavy user of materials such as
steel, non-ferrous metal, rubber, plastics, glass and fibre. Design and production demand
competence in areas as diverse as aerodynamics, fluid dynamics and various engineering
skills. Far from filling an absolute void, web-based technologies are an evolution of
existing technology initiatives — which have often failed to deliver returns in line
with the huge amounts of goodwill, trust and credibility invested by suppliers
(OSAT, 2000). Since the 1960s, systems such as proprietary electronic data interchange
(EDI), a standard format for transmitting business documents between trading partners,
have allowed efficient but only point-to-point communication of large amounts of
data7. Many large companies rely on EDI, but in its traditional form it offers limited
functionality and has not been standardised worldwide. For instance, EDI does not
easily accommodate various paper and envelope sizes. As a consequence, only an
estimated 5 per cent of the value chain (mainly Tier I suppliers) is connected to EDI
(Deutsche Bank and Roland Berger, 2000). Another application is enterprise resource
planning (ERP) software, used most often to automate companies’ internal business
processes such as tracking orders, planning production schedules and managing
inventories. Overall, the lack of high-quality information that can be exchanged among
all players in the industry still produces various forms of inefficiencies. Suppliers and
dealers build huge inventories because they do not have accurate data on their customers’
demand. Delays and associated costs are rampant in vehicle development because
different CAD/CAM proprietary systems are not perfectly integrated and information
has to be delivered manually between parties8.

97
 The diffusion of XML (extensive markup language9) makes it possible to switch
from discrete information and data exchange through EDI to continuous web-based
flows. The real gains from on-line B2B commerce may come not from trading but
from better access to and sharing of information (Berryman and Heck, 2001). This
information might include supply and demand forecasts, reports of inventory levels at
points along the supply chain and market-tested predictions of the effect that the price
of futures and other options will have on the availability of particular supplies, such
as electricity and paper. Significant opportunities for productivity enhancement arise
because networked communication lowers the costs of implementing changes, reduces
the cost threshold of quality improvement, decreases direct and opportunity costs of
communication and decision-making and makes the whole product development cycle
faster and more frequent (Fine and Raff, 2000). Collaborative dealings are made
possible by the development of internet-enabled manufacturing research planning
(MRP) systems to facilitate collaborative product design of complex components or
modules through new engineering, production and training techniques, called
visualisation tools (Sturgeon, 2001). “Building-to-order”, i.e. the customisation of
car design and production according to clients’ needs, is the (problematic) vision of
where the internet may take the car industry in the long term (Helper and
MacDuffie, 2000). The paradigm is the Dell Computer Corporation’s business model
and its zero inventory and retail costs10. In the industry jargon, such additional
opportunities can be classified as:
— E-information management (activities associated with knowledge management,
research and development, and marketing);
— E-documentation (configuration management and maintenance, repair and
overhaul — MRO — activities); and
— E-collaboration (engineering and planning activities).
On the B2B side, the internet is posited as promoting both market-like and
collaborative dealings with suppliers. For the former, arguably the most significant
deal has been the decision of GM, Ford, and DaimlerChrysler to establish the world’s
biggest e-business purchasing joint venture — Covisint (Cupertino, Vision and
Integration) — with two information technology firms, Commerce One and Oracle11.
This is a “many-to-many” B2B supplier exchange on a single global internet portal
designed for supplier auctions and to organise supply chains and plan logistics. The
venture’s core offerings will include services to assist in product design, supply chain
management and procurement functions performed by auto manufacturers and their
direct and indirect suppliers12. Alternatively, “private” on-line exchanges built by
individual companies to connect their suppliers and customers often are complemented
by resort to third parties, who conduct on-line auctions and vet bidders on behalf of
big companies13. Its executives expected Covisint’s growth to be spectacular, but in
the first four months of operation (from October 2000 to January 2001) Covisint saw
only some 100 auctions accounting for about $350 million in products traded. This
was paltry, even in comparison with the private exchange run by Volkswagen, whose
traded volume hovers around $1 billion, let alone the $210-220 billion bought annually
by Covisint’s shareholders14.

98
 Marketplaces have yet to focus on improving business processes to unlock
additional value, since their founders typically focused on the “classical” benefits of
an efficient marketplace: the ability to clear the market quickly and cheaply, to
aggregate the orders of buyers and thus to achieve lower prices. Applications such as
shipper and logistic assignment and tracking are nowhere near implementation for the
Covisint portal — but they are almost routine for established dotcoms. Suppliers,
especially in Tier II and below, have yet to be convinced of the benefits of e-business,
which they see mostly as a tool put in place by OEMs to reduce their margins. Sceptics
also say that most e-commerce forays are driven by nervousness about being left
behind, as opposed to a credible cost-benefit analysis. There is also a feeling that the
auto industry’s current business model is far from appealing, as testified by its lacklustre
stock market performance, and that executives have announced internet initiatives
simply to benefit from the positive effect on equity prices15.
Other than the general conditions needed for internet markets to function,
moreover, car industry portals require a rather radical modification in the way
automobiles are designed and produced, shifting from the integral/closed architecture
to a modular/open one. To accomplish this, components should be standardised across
products and companies, present common interfaces and have specifications widely
known and accessible to a wide range of suppliers. A rather long list of engineering
arguments question the benefits of modularisation, however (Helper and
MacDuffie, 2000 and Sako and Murray, 2000). Moreover, a movement in this direction
must by definition be made collectively; no assembler is willing to act alone. Another
obstacle to the efficient application of internet-based solutions is that, no matter the
speed of information exchange, its accuracy may remain insufficient. The information
might include erroneous production forecasts, faulty release notices, inappropriate
delivery schedules, etc., whether from OEMs or other suppliers. This is a serious
concern because early automation efforts suffered from the rote implementation of
“islands of automation”, nicely automated portions of a process that were then embedded
in an existing system. Such efforts often had little pay-off, sometimes because they
addressed minor portions of the system, sometimes because their apparent
improvements came from simply shifting problems elsewhere in the process. Similar
rote “infomating” of the current system will yield at best little improvement.

Consequences for Market Structure

Fears that exchanges in which rivals come together might encourage collusion
or price fixing have prompted anti-trust regulators on both sides of the Atlantic to
examine B2B exchanges closely — including Covisint. The Federal Trade Commission
opened its investigation of whether the formation of Covisint violates Section 7 of the
Clayton Act in June 2000, concluding in September that “Because Covisint is in the
early stages of its development and has not yet adopted bylaws, operating rules or
terms for participant access, because it is not yet operational, and because its founders
represent such a large share of the automobile market, the Commission cannot say

99
that implementation of the Covisint venture will not cause competitive concerns. In
view of this, the Commission reserved the right to take such further action as the
public interest may require”16. Brussels authorities had provided official support for
B2B ventures of this type in August 2000 when they approved MyAircraft.com, an e-
marketplace serving the aerospace sector established by i2 Technologies, Honeywell,
and United Technologies.
Although cutting through the hype has been rather difficult so far, there is an
emerging trend for OEMs to conduct sourcing of components over the internet. That
has caused considerable unease among suppliers, many of whom fear another attempt
to screw down costs. In principle, insofar as it reduces transaction costs, B2B also
removes entry barriers to smaller suppliers disadvantaged by the high costs of proprietary
IT systems. In theory, nothing prevents a maker of components based in, say,
Bangladesh, to bid to source, say, GM — although of course assets complementary to
e-commerce are not widely available in developing countries (Goldstein and
O’Connor, 2000). In practice, for the local firm in developing countries, a lot will
hinge on the wider industrial organisation implications of e-procurement, particularly
the mode of supplier relations that will dominate (Helper and MacDuffie, 2000).
Reliance on the “voice” model (co-ordination through co-operation), characterised by
a high density of contacts and information flows between the automaker and the
original supplier, may rise if modular design is increasingly outsourced, or if vehicle
design remains integral but the production of OEM-designed components is extensively
subcontracted. In this case, e-procurement won’t proceed by auction and the benefit
from exchanges like Covisint will rather accrue from the provision of timely and
accurate information, in turn aiding co-ordination and collaboration. Firms in the
developing world are unlikely to derive much benefit from such a situation, in view
of their weak capabilities in scheduling and logistics17, poor levels of trust and relational
knowledge and behavioural models hardly conducive to anything more than minor,
incremental innovation.
If, however, modular design advances and is kept vertically integrated, outsourced
production of relatively simple components may reinforce the “exit” model (co-
ordination through the market), where OEMs interact with suppliers by replacing
them when problems emerge.
“The internet can facilitate this model by reducing the transaction costs of
doing business with many suppliers, and with smaller firms that can’t
afford proprietary IT software. It is possible that internet auctions could
lead to a reversal of recent trends toward sourcing from fewer firms, and
from ‘full-service’ firms that can do design and sub-assembly as well as
build to the automakers’ print” (Helper and MacDuffie, 2000, p. 35).
While suppliers in general will likely see their bargaining power reduced, those
that have not invested in design and modular production capabilities (certainly most
firms in developing countries) may find their competitive positions relatively stronger
vis-à-vis competitors with far higher sunk fixed costs.

100
 These factors are likely to have even greater importance in emerging markets,
where the costs associated with e-commerce may be higher than in the assemblers’
home countries, and the benefits of leap-frogging technologies might conversely be
lower. It is also plausible, however, that introducing e-commerce may be easier when
establishing a supply chain from scratch, as is usually the case in emerging markets18.
A parallel can be drawn with the simultaneous implementation of modularisation and
outsourcing in the context of greenfield investment projects in non-traditional automotive
regions, such as Mercedes-Benz’s plant in Alabama19. Finally, in view of the different
propensities of firms to modularise according to their nationality — higher in Europe
and the United States than in Japan (Sako and Murray, 2000) — domestic firms may
derive different benefits according to the origins of their automotive customers.

Recent Trends in the Indian Automotive Industry

The Indian automobile industry has been extensively regulated — in terms of

imports, collaborations, equity ventures, capacity and technology transfer — since
the country’s independence in 1947. In a country characterised by extreme poverty,
government planners considered the passenger car a luxury item and imposed very
high tariffs. Behind the wall of high tariffs, technology was outdated, suppliers were
fragmented and uncompetitive and assemblers maintained an inefficiently high degree
of vertical integration 20. Against the US vehicle population of 760 units per
1 000 people, India stood at a mere ten. Seeing a niche for a “people’s car”, Indira
Gandhi’s son Sanjay set up Maruti Udyog Ltd. (MUL) in 1981. The state-owned firm
soon floundered and was rescued in 1990, when Japan’s Suzuki, the first MNC to
enter the local market, took a minority stake, which in 1992 rose to 50 per cent21. Its
little 800cc car commanded a market share of some 80 per cent in 1996, more even
than its technical annual production capacity of 250 00022. A programme of market
liberalisation was started in 1991 and extended to the car industry two years later,
luring many MNCs to enter the Indian market in various forms (Table 5.1)23.
Although price still largely dominates quality in consumers’ preference criteria,
and new entrants initially imported a large proportion of components in the form of
CKD/SKD kits, more modern manufacturing and organisational routines, such as lean
production, have become increasingly common24. At least initially, the Indian state
has played an important and supportive role in this process of industrial upgrading
(D’Costa, 1995)25. To increase indigenisation26, MUL has largely mimicked the JIT,
TQM, and kaizen strategies of leading Japanese companies (D’Costa, 1995), and has
successfully created linkages with local small and medium-size producers of car
components (Okada, 1998 and 2000), sometimes taking equity positions in them.
Gulyani (2001) documents how Maruti and Ford have tailored lean manufacturing to
the Indian context. She stresses in particular how poor transport infrastructure, by
introducing inefficiencies and unreliability in the supply chain, has created at least
two distinct auto clusters — one in Delhi and another in Chennai — where “assemblers
reduce their vulnerability to and demand for transportation services and therefore
reduce both the direct costs and the external diseconomies associated with poor
transportation systems” (p. 16).

101
 Table 5.1. The Indian Automotive Industry
a
Company Established Foreign Ownership Installed Capacity Sales Employees Models Plants
(percentages)
b
M&M 1945
c
Telco 1945 345 000 74 595 n.a. Indica, Estate, Sumo, Pune
Safari
Maruti 1981 350 000 336 276 5 719 800, Alto, Zen, Omni, Gurgaon (near Delhi)
WagonR, Esteem, Baleno,
Altura, City
Daewoo 1994 91 72 000 47 825 2 319 Cielo, Matiz Delhi
Mercedes 1994 86 9 000 634 338 E-Class, S-Class Pune
d
Peugeot 1994 32 0 -- Closed
Honda 1995 90 30 000 10 051 820 City Pune
Mitsubishi 1996 7 889 Lancer Tiruvallur (near Chennai)
Hyundai 1996 100 120 000 79 795 1 465 Accent, Santro Near Chennai
Toyota 1997 70 50 000 23 576 502 Qualis Bangalore
e
Fiat 1998 100 60 000 10 329 2 352 Siena, Siena Weekend, Kurla (near Mumbai)
102

Uno
e f
GM 1998 100 25 000 7 952 555 Astra , Corsa Halol (near Baroda)
e g
Ford 1999 100 100 000 17 506 608 Escort , Ikon Near Chennai

Notes: a) March 2000-February 2001; b) Mahindra and Mahindra; c) includes vehicles other than passenger cars; d) exited India in 1997; e) as wholly owned subsidiary; f) joint venture
with the Birla group; g) joint venture with the Mahindra group
Source: Society of Indian Automobile Manufacturers (2000), Profile of the Indian Automobile Industry and Autocar India, April 2001.
 At a broader level, Tewari (2001) documents the transformation of the automotive
supply base in Tamil Nadu following the arrival of three international assemblers:
Ford, Hyundai and Mitsubishi. OEMs strove to adopt world-class standards in areas
such as the rationalisation and narrowing of the supply base, focus on full-service and
full-package component delivery. Although many local firms are falling behind, a
fair number of them have been able to upgrade their production systems and compete
with some success27. For example, several suppliers reported bidding for contracts
and receiving requests for quotations (RFQ) from global buyers. Sundaram Fasteners
has even become the world’s biggest suppliers of metal radiator caps to GM (Box
5.2)28. As documented by Narayanan (1998), even in an era of capacity licensing,
development of competitive skills crucially depended upon the ability to build specific
technology trajectory advantages and complement imported technology with in-house
technological efforts.

Box 5.2. The TVS Group29

With sales exceeding $1.7 billion, Chennai-based TVS is India’s largest automotive
component group and one of the country’s largest business conglomerates. Its main
subsidiaries are Sundaram Fasteners Limited (SFL), manufacturing fasteners, cold
extruded parts, powder metal parts, iron powder, radiator caps and gear shifters, and
Sundaram Clayton Ltd. (SCL), producing air brake actuation systems and pressure
die castings. The group is owned by the Srinavasan family and directed by the
founder’s son, an engineer who received his MBA in the United States.
In the mid-1980s, TVS was India’s first company to embrace Total Quality
Management (TQM). With support from two Japanese consultants and a few Indian
business scholars in foreign universities, SCL integrated TQM’s ten core parameters
into its quality practices of total employee involvement, policy deployment,
standardisation, Kaizen (continuous improvement) and training. It set up an ad hoc
quality policy room, where top managers meet at least once a week, and created a
programme to familiarise all staff with statistical tools of quality control (such as the
checksheet, the cause-and-effect diagram and histograms). SCL has 67 quality circles,
some 250 of whose suggestions are implemented every month. By reorganising
product-development teams along cross-functional lines and accompanying every
process by continuous data analysis, time-to-market has been reduced from 24-
30 months to 12-14 months. In 1998, SCL became Asia’s first-ever winner of the
Deming Prize for Company-Wide Quality Control in the Overseas Companies
category, awarded by the Union of Japanese Scientists and Engineers. In the same
year it implemented ERP and in 2000 it put on-line an intranet to combine documents
and data and reduce paperwork.
On the basis of SCL’s success, other TVS companies have achieved world class
competitiveness. SFL was the first Indian company to obtain ISO 9000 Certification,
and it received the Total Productive Maintenance (TPM) Excellence Award given
by the Japanese Institute of Plant Maintenance in 1998. It is a global supplier for
GM, which awarded SFL its best supplier award for the years 1996-99. To satisfy JIT
requirements, SFL has established warehousing arrangements with EDI capabilities
in the United States.

103
 Islands of excellence, however, cannot conceal the generally low competitiveness
of Indian small and medium-sized autoparts makers. Response times are long and
defect rates are on average ten times higher than those in Japan30. Increasing competition
translates into the necessity of reducing time-to-market — and therefore mastering
concurrent engineering — and trimming costs. Consolidation is expected to occur in
coming years, as the most competitive firms get bought by foreign investors and only
a handful of domestic ones survive.
With the introduction of stricter environment norms (“Bharat Stage II”) in 2000
and the phasing out of remaining trade protection in April 200131, auto manufacturers
have started experimenting with the internet (Table 5.2). Improving after-sale service
is emerging as a differentiating strategy and all OEMs maintain B2C sites with general
information on the vehicles and the dealership network. Supply chains distributed
over the vast geography of India could be enhanced by enabling smoother virtual
communication. As summarised in Box 5.3, some large-scale local companies have
already started e-procurement. While India is well known for its fast-growing software
industry (Arora, 2001)32, however, net-based business remains largely uncharted
territory for corporate India. A survey of e-commerce use in 116 Indian companies
found that, while at more than half of them e-commerce constitutes a substantial part
of corporate strategy and internal e-mail is used by all companies, less than a tenth use
the net for accepting customer orders (KPMG, 1999). The mean value of e-commerce
deals, for the quarter of the respondent companies that do track their trades through
the net, amounts to a microscopic 0.025 per cent of average turnover33. The lack of a
standard payments infrastructure and the many suppliers and channels that are not
net-enabled prevent companies from extending their e-commerce links to span the
entire supply chain. While the government is rapidly privatising the phone system,
few of the 28 states support internet connections34. In May 2000 Parliament passed
the IT Act, amending laws to facilitate e-commerce. It provides a legal framework for
electronic contracts and digital signatures, to enable the conclusion of contracts and
creation of rights and obligations through the electronic medium, and to prevent
computer crimes. It allows 100 per cent foreign equity in B2B e-commerce, while no
FDI is allowed in retail e-commerce.

Table 5.2. E-Commerce Initiatives in the Indian Automotive Industry

Initiative Description

All OEMs Websites offer general information on characteristics of vehicles and location of
dealers.
Automartindia.com Second-hand vehicles web site launched by Mahindra and Mahindra.
Autowebex.com Launched by Satyam Infoway, the country’s largest ISP, which also manages
B2B exchanges in such fields as agriculture and pharmaceuticals.
India Auto Exchange Launched in mid-2000. It will include the entire automotive community from
truckmakers to drivers, provide everything from auto parts to mapping services,
and sell cars online.

104
 Box 5.3. MICO Forays into E-commerce
Founded in 1951, MICO is a member of the German Bosch Group and employs
11 000 people. It is the country’s largest manufacturer of diesel fuel-injection
equipment and one of the world’s largest. In November 2000, Trade2Gain.com
completed a private marketplace deal for MICO. Fixed price and dynamic price
auctions for disposal of fuel-injection equipment spares were conducted. The bidders
were 83 MICO distributors across the country. The auctions were a unique combination
of on-line and off-line activities. Various marketing initiatives such as direct mailers
helped bring all the distributors together. The company considers short turnaround
periods, ability to focus on a target audience, quick responses, low costs and less
hassle as the main advantages of auctions.

A Case Study

Fiat’s Global Strategy

At the turn of the 1980s, Fiat was dogged by its high dependence on small, low-
margin cars and on the Italian market. Following the principles detailed in section
two, its strategic response has been to streamline the supply chain, reduce the number
of platforms and form an alliance. To raise productivity and in place of heavy reliance
on automation and robotics in the 1980s, it has emphasised organisational
interdependencies and co-ordinating mechanisms in the last decade (Camuffo and
Volpato, 1994; Buratti, 2000). It has cut the number of direct suppliers from 700 to
340, introduced platforms, simultaneous engineering and process re-engineering and
increasingly involved producers of auto components at early stages in the design and
production cycle (Calabrese, 2000). In the mid-1990s, it opened a new “integrated
factory” at Melfi, in Southern Italy35. Outside the gate is an industrial park of suppliers
of parts, which roll in on electric trucks as the assembly line’s computers order them.
Stocks are counted in minutes and so flexible are the robots and the assembly lines
that new models get phased in without stopping the lines. In the footprints of Melfi,
which now has one of the best productivity records in the world, Fiat has made similar
huge investments in Brazil, Argentina, Turkey and Poland to make its “world car”, the
smart and rugged little Palio (Box 5.4). In 2000, Fiat and GM agreed to exchange
equity participation and to pool some of their activities.

105
 Box 5.4. The Palio36
While most products are launched in manufacturers’ home markets, Fiat developed its
world car in Brazil, to enter emerging markets with an appropriate car designed with
the correct consumers in mind. The car’s origins date back to 1992, when the firm
faced two challenges — to replace the best-selling but ageing Uno in Brazil and to
reduce the firm’s long dependence on the crowded West European market, especially
Italy. Having decided on a new Third-World car, Fiat began to refine its strategy for
“Project 178”, the effort to produce a range of cars that would be modern, sufficiently
flexible to suit local conditions and easy to build in volume. The Palio family includes
three models: a hatchback, the Weekend estate car, launched in Brazil in March 1997,
and the Siena saloon.
The task of designing the car was given to a 300-strong team, which assembled in
Turin during 1993 and 1994. Among them were 120 Brazilians, ranging from engineers
to shop-floor workers, as well as Argentinians, Turks and Poles. This group insisted
that the new car should also have some non-European features37. Although the Palio
broadly resembles the Punto, a success in Europe, it was designed as slightly bigger,
to serve as the sole family car for many of its buyers. The Palio is also stronger and
structurally more robust than the Punto, designed for the rough roads of the Brazilian
interior. For Fiat, whose cars had a long-standing reputation for lightweight tinniness,
this big change meant equipping the car with a higher and more flexible suspension
and using materials appropriate to combat noise, dust and water. Fiat also aimed to
give customers a few frills. Even the basic Palios come with safety features such as
ABS, while optional extras include electronic locking, car alarms and air bags (then a
first for a Brazilian-made car).
During 1994-95, Fiat sent 290 workers and engineers from Betim in Minas Gerais to
Turin to set up a pilot Palio production line and build a prototype. After careful study,
the company decided to automate only those parts of the line at Betim where precision
and quality were vital. The renovated factory was not antiquated. The Palio assembly
line uses rotating cradles that tilt the car body to convenient angles, improving both
quality and job satisfaction. The firm has moved further towards just-in-time
production. Under a previous policy known as minarização, Fiat had encouraged
many of its principal suppliers to set up plants in Minas Gerais. For the Palio, it took
this system further. Eschewing a new factory elsewhere, the firm squeezed the most
out of its investment at Betim, while the Palio’s huge potential production volume
allowed Fiat to negotiate worldwide contracts with its chief suppliers. It persuaded
some, such as Britain’s BTR, to make big investments in developing components for
the car and building new factories for them. Compared with around 200 component
suppliers for the Uno, the Palio has just 130, the most important with factories close to
Betim and some 40 linked by computer to Fiat’s assembly line. Dealer groups have
been cut from 420 to 335, of which the 248 main ones are connected by a satellite link
so that customers can specify the accessories they want and get the finished product
about 30 days later.

106
 Fiat’s Strategy in India

At the start of the 1990s, Fiat sold only 140 000 cars outside Europe; in 2000 it
sold more than 800 000. The Palio has given Fiat the lead over Volkswagen in the
Brazilian market, which accounts for 70 per cent of all South American car sales. The
main plank of the strategy now is to build as many cars as possible on the same basic
chassis to yield the economies of scale enjoyed only by the likes of Toyota, with its
Corolla model. Although the Palio family model is produced in ten countries, and
more than 1.5 million cars have been sold to customers in 41 countries in four years,
Brazil still accounts for 80 per cent of total production38. The Palio family was designed
to be flexible, and details such as colour schemes and manufacturing techniques can
be adjusted to suit local needs. Brazil is a rather sophisticated market as emerging
economies go. In India, well-off consumers who can buy a car can often also afford a
chauffeur to drive it, so the specifications are different. For example, electric windows
are in the rear rather than the front (Humphrey, 1999, p. 9) and there is more legroom
in the back.
Fiat has been present in India indirectly since 1944 through a licensing agreement
with Premier Automobiles Ltd. (PAL), owned by the Walchand group of the Doshi
family39. In the 1960s PAL started production of the Padmini, an indigenised version
of the glorious Fiat 1100 launched in Europe in 1962 and still in production in the
early 1990s. The PAL factory at Kurla in Northeast Mumbai, however, was a typical
baniya operation, milking the assets without major reinvestment and retooling
(D’Costa, 2000). When the local market opened in the 1990s, Fiat decided to follow
a dual strategy. First, it granted a licence to PAL to assemble CKD Uno kits in Kurla
and set up IAL, a joint venture. Second, Fiat India Automobiles Ltd. (FIAL), a wholly
owned subsidiary, signed a memorandum of understanding with the Maharashtra
government in 1997 to build a plant in Ranjangaon, 60 km from Pune. The project
involved transplanting the modular supplier park to India, with suppliers building
their own dedicated facilities close to the OEM to produce modules
(Magnabosco, 1999), and sourcing 65 per cent of the Palio’s components locally by
1999 and 80 per cent by 200040.
Introduced in India in 1996, the Uno generated much interest. Over
290 000 bookings were received, but production problems quickly emerged and
dissatisfied customers cancelled their orders41. Poor management and internal family
squabbles made it impossible for the Doshis to put any more cash into the business.
Fiat was compelled to refund to customers who had cancelled their orders, acquire
majority control in the Kurla operations and incrementally increase its shareholding
in IAL to the current 93 per cent42. The Uno sold around 1 800 units per month in
1999 but, although both the petrol and diesel versions had long waiting lists, the
company could not increase domestic content and suffered from poor marketing and
sales efforts43. Fiat came to the conclusion that it made little sense to pursue its plan
for a separate greenfield facility in Ranjangaon, as there was adequate capacity at the
Kurla plant to make the Palio Siena.

107
 Fiat has invested $600 million in India, completely revamping the Kurla plant,
especially to improve the welding and integrated paint shops. Domestic content of the
Siena rose from 30 per cent at the time of launch to 60 per cent by end-2000. Given
the low cost of the Indian workforce, the firm has not installed robots, even in the
stamping area and the paint-shop. Fiat brought down manpower at Kurla, famous for
labour unrest, by a little over 1 000 and reduced the number of hierarchical levels to
seven. An internal union backed by management has overtaken the traditional unions.
Team leaders, or supervisors, were appointed, and about 80 of them were sent for a
month to the Fiat plant in Turkey44. Despite the introduction at Kurla of one of the
basic organisational features of the integrated factory model — the so-called elementary
technological units (UTE) — fully implementing lean manufacturing has been very
difficult. Because of labour market regulations, Fiat has been unable to substitute a
younger workforce for PAL workers and has found it hard if not impossible to adopt
concepts such as continuous improvement, total employment involvement and quality
control circles, and to change the value system45. Logistics posed further limits. Kurla
is a dense urban area of Mumbai, with very high estate prices although slums surround
it. As a result (and also because production volumes have remained rather modest) the
scope for realising JIT has proved limited. Table 5.3 shows the geographical distribution
of Fiat’s suppliers. While those located in the Mumbai area represent more than a
third of the total and supply more than half of total items, they account for only
10 per cent of the car’s total value. The most important suppliers are located around
other OEMs in the Delhi, Pune and Chennai auto clusters, although transport takes
between one and five days. Because of poor logistics, the inventory is kept at three
days. Another cost associated with a geographically spread supply chain is that the tax
burden is 2 to 3 per cent higher in Mumbai (especially on account of the octroi, an
entry tax charged by the municipality of final destination).

Table 5.3. Fiat’s Indian Suppliers

Cluster Distance from Kurla Transit Time Number of Number of Items Percentage of
(kms) (days) Suppliers Total Value
Delhi 1 408 5 29 124 34
Chennai 1 367 5 11 53 19
Bangalore 1 033 4 7 22 3
Nagpur 861 3 2 6 0
Indore 600 3 1 1 3
Baroda 500 2 1 9 9
Aurangabad 400 2 1 1 3
Pune 200 1 25 146 20
Mumbai 100 1 43 386 10

After struggling in 1999-2000, Fiat completely restructured its 50-dealer network,

which was set to go up to 90 by year-end 2001.The launch of the full Palio line in
late 2001 is expected to raise Fiat’s market share to 15 per cent by 2005. On the back
of strong commonality with the Siena version, the other Palio models are expected to
roll out with a localisation of 70 per cent and to allow full use of the Kurla assembly
line. If the new cars prove a success, Kurla could be decommissioned and Ranjangaon

108
finally opened. The company also intends an increase in the export of components
and is considering the possibility of developing India as a competitive pole for global
sourcing of skins and metal sheets.

Box 5.5 Helping Local SMEs to Grow46

Magneti Marelli and UNIDO launched a Partnership Programme in 1999 aimed at
raising the production capability of Indian auto parts suppliers in the western region
of the country. In its initial phase, the Programme targeted 20 small and medium-
sized companies for non-capital changes induced through intensive shop-floor
interventions, seminars and workshops on manufacturing methods, and marketing
and study tours. The expert assistance was delivered in three rounds with a minimum
of three weeks between them to allow enterprises to assimilate lessons learned and
adapt them. The shop-floor interventions were supported by a workshop on lean
manufacturing methods and marketing, plus three study tours (two national and one
international) designed to help augment the knowledge gained. Early results in
increased productivity, cleaner production methods, better use of technology and
growing awareness of the need for continuous improvement appear to be positive.
During the first demonstration phase, over just nine months, the average lead-time
required for product completion was reduced by 52 per cent. The amount of shop
floor training grew from near zero to 238 hours per month. Absenteeism dropped by
39 per cent. The application of standard operating procedures rose by 54 per cent,
and production space grew by 25 per cent with the introduction of single-flow lines
and better control of stocks, waste and scrap. The UNIDO-Fiat programme is considered
one of the best projects currently running as a test case of involving multinationals to
help in development.

The Internet and Supply-Chain Management at Fiat India

Like other global carmakers, Fiat has launched a number of e-business initiatives
and expects to derive one third of its revenue from services offered over the internet by
2003 (Fusignani, 2001). More than a dozen different initiatives have also been launched
to optimise supply-chain management, enhance design collaboration with other firms
and improve internal operations related to knowledge and human resources management
(Table 5.4)47. Fiat India has made a substantial ITC effort since 1998, implementing
Baan, an enterprise resource planning (ERP) package, and successively introducing
applications such as material requirements planning, EDI, manufacturing execution
systems, CAD, customer relationship management, intranet48, and, in 2001, the internet.
Not surprisingly, Fiat does not operate solutions such as supply-chain management
and enterprise application integration, which are standard requirements for e-commerce
in OECD countries. Individual internal operating systems differ markedly in their
degree of integration with external electronic networks. In design49 and production, e-
mail is used to exchange specifications for components with suppliers, although Fiat’s
increasing reliance on co-design arrangements means that drawings cannot always be
transferred. While most suppliers are equipped with CAD systems such as Computer
Vision, Fiat operates CATIA workstations, so communication has to be translated into
other languages such as I-deas. Tellingly, in Italy all suppliers have CATIA50.

109
 Table 5.4. Fiat Group’s Projects in E-Business

Objective Initiative Firm(s) involved

Optimise the supply-chain

Dealerships Spare parts (on-line orders for repair shops) Fiat Auto
Health Check Dealer Management System (network to Fiat Auto
connect all dealers)
Internet SAP Ferrari
e-CRM (integrated offer for dealers) Iveco
Suppliers e-Procurement (direct portal) Fiat Group (open to
Phase I (under launch): commodities, indirect non-Fiat)
materials, IT
Phase II (project in progress): direct components
Fast Buyer Magneti Marelli
Improve internal processes CWW-MYFIAT (knowledge management portal) Fiat Auto
HR portal (management of HR processes) Iveco
Lengthen the supply chain Worldwide Financing (on-line sale finance) Fiat Auto
Fidis fornitori on-line (home factoring) Fidis
Savarent on-line (fleet administration) Sava
Targa.com (assistance, infotainment) Fiat Auto
Generate value Fast Buyer (portal for e-procurement of raw materials) Business Solutions
Ciao Web (generic portal) Fiat and IFIL
OnTruck.com (vertical marketplace) Iveco

Source: Fiat.

In procurement, the internet is hardly used, although e-mail serves for negotiation
and billing. A new system has been introduced, however, to send material requisition
planning orders to suppliers by e-mail, with potentially high savings on phone calls51.
Payment is supported on EDI. Fiat expects its suppliers to have e-mail addresses, if
possible one for sales and one for production, but does not require it — contrary to what
seems to be the case at Maruti52. Aggregated information on the degree of e-awareness
among suppliers shows a high correlation between foreign ownership, quality, size and
e-mail availability. Suppliers in the Mumbai area, in particular, are smaller, produce
goods of relatively lower quality and are less likely to have their own internet sites53.
Fiat India has never considered procuring through auctions, although management
expects no less than 80 per cent of its purchasing operations in India to be put on the
internet by 2005. Lack of convincing evidence on the real benefits of B2B combines
with the perception that, in India at least, business confidence has to be built over time
through face-to-face interaction. An interesting finding is that one supplier — admittedly
a member of India’s largest business group and probably the first local firm to participate
in an automotive auction — submitted a bid for a component on a Europe-based site,
and to management’s surprise discovered that its price is internationally competitive.
This company thinks that assemblers’ reluctance to share more information with
potential suppliers is a limit on web bidding, since drawings are available only as
image files and do not allow a precise understanding of components’ specifications54.

110
 All white-collar employees have intranet and e-mail, although only managers
have internet access. Not surprisingly, e-business is having a much greater impact on
the internal organisation of the firm, although probably at a slower pace than in other
Fiat subsidiaries, partly reflecting low production volumes. In procurement, for
example, process audits of suppliers by the quality team go on the intranet and are
accessible in other subsidiaries as well through a dedicated bulletin board (bacheca di
localizzazione). Imports of knocked-out kits of complex systems are electronically
co-ordinated through the company’s headquarters in Italy, and Fiat India will eventually
become a full participant in the Global Sourcing System, Fiat’s private exchange.
Shipments, orders, and operational plans can be traced in real time through the World
Material Flow System, although follow-up calls are common. In marketing, Fiat put
a considerable emphasis on the development of the internet ahead of the Palio launch.
The nine zonal representatives are armed with a mobile phone and a laptop to ensure
that they remain effective while roaming among the dealer outlets in the country. The
dealers are connected to Fiat through the dealer on-line system (eDos), which allows
full sharing of information on stock levels. More generally, communication flows
with the headquarters and other subsidiaries are now conducted mostly through e-
mail, and the intranet is used for trouble-shooting, often on the basis of previous
experiences in other foreign plants. A starting-kit pool of application systems,
aggregating legacy systems in use elsewhere in the Fiat universe and some India-
specific applications, which was developed a few years ago, is now proposed for the
new Fiat subsidiary in Thailand.

Conclusion

This paper embarked on an analysis of the use of e-business in the Indian car
industry, taking Fiat India as a representative case study and looking in particular at
the effects on (domestic) companies in the supply chain. As the research progressed, it
became clear that the focus had to be broader. While net-based solutions are expected
to streamline the automotive supply chain, they are necessarily of lesser importance
than progress in implementing lean manufacturing. In this sense the choice of Fiat
India may not be optimal for the present research because the company has struggled,
for different reasons detailed above, to put supply-chain management principles into
operation. At the same time, Fiat represents an ideal case study because it produces a
homogeneous product for emerging markets in eight countries, including India, and
has been very successful in optimising supply-chain management in Brazil, where it
operates one of the largest non-OECD car factories55. Moreover, and despite much
lower production volumes, there is no reason to expect Fiat India to be less internet-
savvy than its local competitors. The other contextual element of great importance is
the upgrading of local firms, SMEs in particular. Here again, the focus on a single
company has allowed access to a rare source of information, and suppliers at any rate
also sell to other OEMs.

111
 The broad picture emerging from this research is one of deep transformations.
The auto business in India, and supplier-OEM relationships in particular, are not
immune from global pressures, including over-supply, industry consolidation, modular
assembly and the increasing use of the internet. Internet use now appears to be a core
element in the internal organisation of OEMs, a fortiori of a global company like
Fiat, and to be taking an increasingly important role in organising inter-firm
relationships. On the other hand the use of the medium is still very limited in intra-
firm relationships. Such effects on the supply chain are clearly more tangible in terms
of e-information management (knowledge management, research and development
and marketing) than in procurement. These results appear to be in line with those of
other studies (e.g. Hawkings and Prencipe, 2000). There also seems to be a high
correlation between size and digital awareness, with SMEs around the world showing
a lower propensity to using the internet. In short, almost regardless of the industry
and the geographical region, early research has hinted that the impact is still minor.
Companies are not putting in place necessary measures to reap potential cost savings
and efficiency improvements, and neither customers nor suppliers are applying real
pressure — despite strong theoretical arguments that the internet allows efficiency
improvement by exchanging information and developing collaborative approaches all
along the supply chain56.
From a development perspective, one disquieting conclusion is that the much-
awaited rebirth of local components manufacturers on the basis of lower entry barriers
is still well down the road. While Fiat may not be able fully to achieve the efficiency
goals of lean manufacturing, its suppliers are certainly representative of a wider
universe. Hard data show that their insertion into global supply chains has hardly
augmented in recent years, and evidence gathered during the fieldwork confirms that
small-scale suppliers do not export through the internet. That the only firms credited
with bidding on the internet belong to diversified business groups confirms that they
still represent the backbone of industry in emerging markets (Khanna, 2000). It remains
to be seen whether the fact that e-business does not modify relative bargaining positions
in the automotive supply chain reflects country-specific or industry-specific peculiarities.
On the other hand, in view of the evolutionary nature of e-business and of the
impact it is having on corporate organisation, the scholarly and policy implications
are probably much clearer. Research should focus on how net-based applications change
organisation routines, modify the pool of corporate resources and interact with strategy
in firms of different size, nationality, ownership and industry. Given the lack of
knowledge about the drivers of corporate growth, the value of aggregate analysis is
probably very low. Policy makers should aim at solutions capable of increasing the
general use of ICT and the internet, starting with the removal of existing barriers, and
refrain if possible from targeted initiatives on e-commerce whose value is limited, at
least in the short term. The upgrading programme in Tamil Nadu run by Fiat, Magneti,
Marelli and UNIDO does indeed support the conclusion of Tewari (2001) that
“innovative programs of support by the government could go a long way in diffusing
‘successful’ processes more widely”.

112
 Notes

1. See Goldstein and O’Connor (2000) on the implications of e-commerce in the

productive sectors of developing countries.
2. The focus is on the interaction between corporations along the supply chain rather
than on the implications of “clicks” for “bricks” — the so-called business-to-
consumers (B2C) segment. One of the earliest studies on internet car retailing in the
United States finds a rather minor, although positive, effect on prices: buyers who
use an internet referral service save about 2 per cent on their average new car purchase
(Scott Morton et al., 2000).
3. According to US Commerce Department figures reported in “E-business to Business
Soars”, Financial Times, 8 March 2001.
4. The propensity to source components from local firms is also a function of the
extent to which adaptations are made for the local market and the resource intensity
of the parts involved (Barnes and Kaplinsky, 2000).
5. This section tries to summarise in a few paragraphs a body of literature that could fill
various storehouses, so the author asks forgiveness for being selective if not
inaccurate.
6. Similar dynamics are clear in other industries producing complex systems, such as
aircraft, with manufacturers of components intervening as risk-sharing partners.
7. EDI requires users to purchase a value-added network (VAN) connection using
leased lines.
8. In the case of Fiat, for example, Calabrese (2000) reports that 70-80 per cent of
communication among people involved in product development takes place through
face-to-face interaction, and only 10 per cent is via computer.
9. A mark-up language is a tool to identify structures in a document. Structured
information contains both content (words, pictures, etc.) and some indication of what
role that content plays (for example, content in a section heading has a different
meaning from content in a footnote, which means something different from content in
a figure caption or content in a database table, etc.). Almost all documents have some
structure. The XML specification defines a standard way to add mark-up to documents.
10. Precisely because Dell Computer derives a competitive advantage from its exclusive
collaborations and from the proprietary sharing of information with its suppliers, it
has avoided public B2B marketplaces and exchanges (Cohen and Agrawal, 2000).

113
11. Renault-Nissan, Toyota and Peugeot joined later. Elsewhere in the motor sector a
variety of B2B operations are being established. Partstrade.com is a UK-based site
e-marketplace that brings together companies in all areas of the aftermarket. Buyers
are able to request quotes for parts and services which are distributed to
Partstrade.com’s network of registered suppliers. Another area of B2B potential is
seen in the procurement of raw materials. A third example is Bluecycle.com, a B2B
site supported by the insurance group CGNU that offers auto salvage that has been
cleared through the insurance claims process.
12. Independent marketplaces also exist that charge a small fee for matching up buyers
and sellers. Most of them, however, failed to generate enough liquidity to become
sustainable ventures.
13. GM had previously created TradeXchange and Ford had set up AutoExchange.
14. “Big Business Gets to Grips with Net Savings”, Financial Times, 6 March 2001.
15. In late 1999, OEMs’ total market capitalisation was one-third of their total turnover,
while it was 22 times Microsoft’s annual sales.
16. See http://www.ftc.gov/opa/2000/09/covisint.htm. A buyer-dominated B2B
exchange need not overtly fix prices in order to engineer the results it wants. A
dynamic price discovery mechanism — in which the buyer chooses a reverse auction,
declares up front the desired quantity of a component to procure and makes suppliers
compete in a downward price spiral — leads to a rent transfer with no need of price
fixing.
17. However, insofar as the quality of developing countries’ physical infrastructure
remains low, OEMs may find it difficult to reduce rush orders by increased sourcing
via the web. Treacherous roads and poor communications infrastructure may
constitute an incentive for the adoption of modular forms of production organisation.
18. As Fine and Raff (2000) put it, “present practices and possibilities are often partly
determined and highly constrained by past choices and historical events” (pp. 3-4).
19. Volkswagen, for instance, is developing its Net 2000 e-procurement system in Brazil
in view of its later adoption in the rest of the world (“Metamorfose Digital”, Exame,
6 September 2000). It is interesting to note that many OEMs experiment in Brazil
with some of the world’s most advanced modular assembly plants, such as VW’s
Resende and GM’s Gravatai.
20. In other words, they chose to “make” rather than “buy”.
21. Under the terms of the joint venture, Suzuki and the government take turns in
nominating MUL’s managing director, for five years at a time.
22. “Local hero”, The Economist, 14 August 1997.
23. Under the present norms, foreign car manufacturers have to invest a minimum of
$50 million over three years, increase domestic content to 50 per cent by the third
year and 70 per cent over five years, and attain minimum levels of export earnings.
24. Other factors that have made upgrading necessary have included obligations to
install catalytic converters, seat belts and laminated safety glass.

114
25. Relationships between the Indian government and Suzuki, however, have become
increasingly bitter. In 2000, Suzuki agreed in principle to the government’s
divesting its 50 per cent stake and expressed its desire to see this stake sold to the
public. The move might not find favour with the government, as it was unlikely to
fetch the desired price. Suzuki is believed to have asked the government for a veto
power on the possible sale of the latter’s stake to a new partner.
26. By 1993, the Japanese content of Maruti fell from over 90 per cent to 5-6 per cent
(“Cars by the million”, Financial Times, 30 September 1993).
27. In 1999, roughly half of the 360 largest suppliers had ISO 9000 certification, although
only 20 also had QS 9000-status (DREE, 1999).
28. Similar stalwarts exist in other emerging markets, such as Brazil’s Sabó Retentores
— a global supplier of oil rings, rubber hoses, and gaskets for Volkswagen — or San
Luís — a Mexican firm that supplies suspension systems to Detroit’s Big Three.
29. This box is based on “Total Quality Ltd.”, Business Today, 22 November 1998.
30. According to a study by A.T. Kearney cited in “Steering in uncertain territory”,
Business Standard, 21 November 2000.
31. At present, auto firms have to balance the actual c.i.f. (cost including freight) value
of imports of SKD/CKD kits and components and the f.o.b. (freight on board) value
of exports of cars and components. The United States, the European Union and
Japan had said this condition violated World Trade Organisation (WTO) rules. The
European Union had even filed a petition before the dispute settlement body of the
WTO. Foreign exchange neutrality norms have not applied since 31 March 2001,
when all QRs were lifted.
32. DaimlerChrysler set up a 50-strong Research Centre-India in Bangalore, which
focuses on encryption technology for security and communication and passive
sensing mechanisms for detection of traffic and road conditions. Delphi, a global
Tier-1 supplier, inaugurated a software development centre in Bangalore in
November 2000, to programme software for engine-control computers. Fiat Group
company Comau has a facility in Pune to source software elements of project
engineering.
33. In the fourth quarter of 1999, e-commerce accounted for 0.6 per cent of US retail
sales.
34. There are just 2.6 phone lines for every 100 Indians, compared with a world average
of 15 per 100. Existing bandwidth meets only 60 per cent of current demand. India’s
first set of private ISPs, ending the three-year monopoly of the government-run
telecom operators Videsh Sanchar Nigam Limited (VSNL) and the Department of
Telecom (DoT), were set up in 2000. There are no limitations to the number of
allowable licences, licence fees would be levied for the first five years, and only a
token fee of about two cents a year would be due from the ISPs over the first ten
years. More than 175 licences have already been issued to ISPs operating in national,
regional and city-specific areas (see “Startup Fever in India”, Siliconindia,
February 2000).

115
35. See the proceedings of the International Workshop “Lean Production and Labour
Force in the Car Industry” held at the Università della Calabria on 25-27 March 2000
(http://www.sociologia.unical.it/convdottorati.html).
36. This box is based on “A car is born”, The Economist, 11 September 1997, Borges et
al. (2000), and “Fiat cruises along Brazil’s difficult roads”, Financial Times,
4 April 2001.
37. Over the same period Ford invested $1.1 billion in a new assembly line at its factory
in São Bernardo, a São Paulo suburb, installing scores of expensive robots. Its aim
was to produce exactly the same car, in exactly the same way, as it does in Western
Europe. But the launch was troubled: delays and two recalls because of minor
defects dented Ford’s Brazilian sales and profits in 1996.
38. “Brazil Launch for New World Car”, Made in Fiat, October/November 2000.
39. Fiat’s entry into India occurred around 1906 via an ambitious Englishman, D’Arcy
Baker, who organised bus rides for journalists around Bombay at a then illegal
14 mph, and a race for the first taxi drivers. Between 1929 and 1934, around
800 Fiat 509 cars, commercial chassis, and taxicabs were imported into India. Fiat’s
commercial vehicle subsidiary, Iveco, is also present through Ashok Leyland’s engine
plant, a joint venture with the Chennai-based company. See “The Indian Connection:
90 years of Fiat in the Sub-Continent”, Made in Fiat, August/September 1999.
40. “Fiat India Signs MoU for Car Project in Maharashtra,” Financial Daily,
7 December 1997.
41. PAL also set up a plant in Kalyan in Northeast Mumbai with Peugeot, an equally
beleaguered joint venture.
42. The firm’s name changed to Fiat India Ltd. (FIL) in September 2000, although
members of the Doshi family still hold management responsibilities.
43. The Uno’s indigenisation level has risen from 52 per cent in 1997 to 70-plus per
cent now.
44. “Italian Lady Injects Fresh Lease of Life into PAL Kurla Plant Work Force”, Financial
Express, 1 March 1999.
45. Fiat is indeed the only OEM that does not operate from a greenfield; Ford started
with a joint venture but, after meeting initial resistance in putting new organisational
measures in force, decided to move to southern India.
46. This box is based on “Impact of the UNIDO/FIAT Partnership”, 10 February 2000
(http://www.unido.org/doc/f330871.htmls) and “UNIDO Embarks on a Policy Shift”,
The Hindu, 2 November 2000.
47. In Italy it entered internet sales with a “dealer-centric” OBS (on-line buying service)
approach in June 2000, first with the Fiat brand (Buy@Fiat), adding Alfa and Lancia
in September. In October 2000, Fiat and Alfa opened web sales sites in the UK and in
November used cars were added in Italy. See “Testore details Fiat’s future with GM”,
Automotive News Europe, 20 November 2000.

116
48. Intranets are designed to give a group of trusted employees access to a set of resources
within one organisation. Fiat India does not have an extranet, i.e. a private business
community that uses the internet as the backbone to extend mission-critical resources
to partners, suppliers, customers and other individuals outside the physical walls of
an organisation.
49. Fiat India obviously does not design new cars, but rather adapts, restyles, and remodels
existing ones.
50. Other firms employ a similar set of ICT tools. Maruti in particular has a rudimentary
MRP, uses EDI in dealing with suppliers, does not trade on-line and enjoys no
inventory visibility along the production pipeline. Interview with Arindam
Bhattacharya, A.T. Kearney, 19 April 2001.
51. Managers report that telecom prices in India are very high and that excessive red
tape has prevented Fiat from maximising the use of dedicated phone lines to
communicate with Italy.
52. Interview with Krishnan Arora, VP Purchasing, Fiat India, 16 April 2001.
53. As predicted on the basis of India’s comparative advantage in electronic engineering,
Fiat already outsources work in 3D modelling, although not for simulation or software
development.
54. Interview with Raghu Mayya, Tata Auto Plastic Systems, 19 April 2001.
55. Not to mention that, for a variety of reasons, other OEMs could not prove as co-
operative as Fiat.
56. In India these savings are estimated to total $400m (interview with Clifford Patrao,
PricewaterhouseCoopers, 19 April 2001).

117
 Bibliography

ARORA, A. (2001), “The Globalization of Software: The Indian Software Industry”, presented
at the Sloan Globalization Conference, Napa Valley, February.
BARNES, J. (1999), “Globalisation and Change: Major Trends in the International Automotive
Industry and their Likely Impact on South African Assembly and Components
Manufacturers”, School of Development Studies Working Paper 23, University of
Natal, Durban.
BARNES, J. and R. KAPLINSKY (2000), “Globalisation and the Death of the Local Firm? The
Automobile Components Sector in South Africa”, Regional Studies, Vol. 34, No. 9.
BERRYMAN, K. and S. HECK (2001), “Is the Third Time the Charm of B2B?”, The McKinsey
Quarterly, No. 2
BORGES, L., C. MAURO, C. DINIZ, F. BORGES TEIXEIRA DOS SANTOS, M.A. CROCCO and O. CAMARGO
(2000), “O Arranjo Produtivo da Rede Fiat de Fornecedores”, Nota Técnicas 17,
Project “Arranjos e Sistemas Produtivos Locais e as Novas Políticas de
Desenvolvimento Industrial e Tecnológico”, UFRJ, Rio de Janeiro.
BURATTI, N. (2000), “New Product Development as Knowledge Management: How Many
Goals Have Been Scored?”, prepared for the 5th International Conference on
Competence-Based Management, Helsinki, 10-14 June.
CAMUFFO, A. and G. VOLPATO (1994), “Making Manufacturing Lean in the Italian Automobile
Industry: the Trajectory of Fiat”, Actes du GERPISA, No. 10.
CALABRESE, G. (2000), “Reorganising the Product and Process Development of an Italian
Car Manufacturer”, in U. JURGENS (ed.), New Product Development and Production
Networks, Springer-Verlag, Berlin.
COHEN, M. and V. AGRAWAL (2000), “All Change in the Second Supply Chain Revolution”,
Financial Times, Mastering Management Supplement, 2 October.
D’C OSTA , A. (2000), “Capitalist Maturity and Corporate Responses to Economic
Liberalization in India: the Steel, Auto, and Software Sectors”, Contemporary South
Asia, Vol. 9, No. 2.
D’COSTA, A. (1995), “The Restructuring of the Indian Automobile Industry: Indian State
and Japanese Capital”, World Development, Vol. 23, No. 3.
DEUTSCHE BANK and ROLAND BERGER (2000), Automotive e-Commerce. A (Virtual) Reality
Check.

118
DREE (1999), “Le marché des composants automobiles en Inde”, mimeo, Embassy of
France in India.
FINE, C. and D. RAFF (2000), “Internet-Driven Innovation and Economic Performance in
the American Automobile Industry”, prepared for the conference “The E-Business
Transformation”, Washington, 26-27 September.
FUSIGNANI, F. (2001), “Esperienze Fiat”, prepared for the conference “Next Business”, Turin,
23 January.
GOLDSTEIN, A. and D. O’CONNOR (2000), E-Commerce for Development: Prospects and Policy
Issues, Development Centre Technical Papers No. 164, OECD, Paris.
GULYANI, S. (2001), “Effects of Poor Transportation on Lean Production and Industrial
Clustering: Evidence from the Indian Auto Industry”, World Development. Vol. 29,
No. 7.
HAWKINS, R. and A. PRENCIPE (2000), “Business-to-Business E-Commerce in the UK: A
Synthesis of Sector Reports Commissioned by the Department of Trade and Industry”,
mimeo, SPRU, University of Sussex.
HELPER, S. and J. MACDUFFIE (2000), “E-volving the Auto Industry: E-Commerce Effects on
Consumer and Supplier Relationships”, prepared for the conference “E-Business
and the Changing Terms of Competition”, Berkeley, 12 September.
HUMPHREY, J. (1999), “Globalisation and Supply Chain Networks: the Auto Industry in
Brazil and India”, in G. GEREFFI, F. PALPACUER and A. PARISOTTO (eds.), Global Production
and Local Jobs, International Institute for Labour Studies, Geneva.
KHANNA, T. (2000), “Business Groups and Social Welfare in Emerging Markets: Existing
Evidence and Unanswered Questions”, European Economic Review, Vol. 44,
No. 4-6.
KPMG (1999), E-Commerce Survey of Corporate India.
LUCKING-REILEY, D. and D. SPULBER (2001), “Business-to-Business Electronic Commerce”,
Journal of Economic Perspectives, Vol. 15, No. 1.
MAGNABOSCO, M. (1999), “Dalla fabbrica integrata alla fabbrica modulare: le nuove frontiere
competitive della Fiat Auto”, in G. SIVINI (ed.), Oltre Melfi, Rubattino.
MUKHERJEE , A. and T. S ASTRY (1996), “Automotive Industry in Emerging Economies”,
Economic and Political Weekly, 30 November.
NARAYANAN, K. (1998), “Technology Acquisition, De-Regulation and Competitiveness: a
Study of Indian Automobile Industry”, Research Policy, Vol. 27, No. 2.
OECD (2000), “Summary of the Methodology for Accessing the Dynamics and Impacts of
Electronic Commerce”, WPIE Ad Hoc Technical Expert Group, Paris.
O KADA, A. (2000), Workers’ Learning through Inter-firm Linkages in the Process of
Globalization: Lessons from the Indian Automobile Industry, Doctoral dissertation,
Department of Urban Studies and Planning, MIT.
OKADA, A. (1998), “Does Globalization Improve Employment and the Quality of Jobs in
India? A Case from the Automobile Industry”, mimeo, IMVP-MIT.

119
OSAT (Office for the Study of Automotive Transportation) (2000), The Automobile Industry:
Moving @ eSpeed, University of Michigan Transportation Research Institute.
SAKO, M. and F. MURRAY (2000), “Modules in Design, Production and Use: Implications for
the Global Automotive Industry”, mimeo, Said Business School, University of Oxford.
SCOTT MORTON, F., F. ZETTLEMEYER and J. SILVA RISSO (2000), “Internet Car Retailing”, NBER
Working Paper 7961, National Bureau of Economic Research, Cambridge, MA.
SDS (School of Development Studies) (2000), “Supply Chain Management: Assisting
South African Industry to Upgrade”, Policy Brief 3, University of Natal, Durban.
STURGEON, T. (2000), “How Do We Define Value Chains and Production Networks?”, IPC
Working Paper 01-005, MIT, Cambridge, MA.
TEWARI, M. (2001), “Engaging the New Global Interlocutors. Foreign Direct Investment
and the Re-shaping of Local Productive Capabilities in Tamil Nadu’s Automotive
Supply Sector”, mimeo, Harvard University, Center for International Development,
India Program.
UNCTAD (2000), World Investment Report, Geneva.

120
 Chapter 6

The World’s First Internet Coffee Auction:

Design, Implementation and Lessons Learned
Morten Scholer

Behind every success lie hard work, substantial risks, sleepless nights and — not
least — the bit of luck that can make the whole difference. This brief paper looks at
elements that turned into a success the world’s first internet coffee auction, held in
December 1999. Lessons are described here for others to learn, but also to avoid false
expectations among the many who say, “Getting my coffee on the internet would
change everything — for the better”.

The Internet Auction of Brazilian Coffees

Coffee people still talk about the Brazilian coffees that were offered
simultaneously on four continents. The auction was a success in at least two ways:
— the technology worked; and
— the coffees sold at prices substantially above expectations.
In October 1999, 315 coffees from different regions in Brazil participated in
the “Best of Brazil” coffee quality contest. In the final round with 24 coffees, a group
of internationally recognised coffee experts, so-called “cuppers”, selected ten coffees
to be offered at the auction. A total of 900 bags of 60 kg each were made available for
sale. A Brazilian exporter was appointed to represent all ten participating farms. The
auction was announced on a web site well known within the coffee trade, that of the
Specialty Coffee Association of America (SCAA) (www.scaa.org). SCAA’s staff and
their webmaster helped prepare the live website that would be used for the auction.
Twenty-three applicants qualified as bidders. They received samples of all the coffees,
detailed instructions on the event and a password for on-line participation. In the days

121
leading up to the auction, they had the opportunity to participate in a trial bidding just
to get familiar with the system — sitting at their screens in the United States, Brazil,
Europe and Japan. The auction started on 15 December 1999 in the American morning
and lasted for 48 hours. The ten coffees were introduced one by one, every five
minutes. The auction also closed coffee by coffee every five minutes, 48 hours later.
To no surprise, the real “fight” for each of the coffees took place during the last hour
of the auction.
Brazilian coffees normally sell at prices below the New York C-price, a common
benchmark in the coffee trade. The auction coffees attracted an average price of
$1.73 per lb. at a time when the New York C-price was approximately $1.00 per lb.
Two coffees went at prices above $2.00 per lb.

Why Sell Coffee via the Internet?

The idea of selling coffee at an internet auction developed from two independent
rationales. The first turned out to be right, but the second was only partially vindicated.
The two notions were:
— a wish to see the spirit of competition among the coffee growers extended to
subsequent links in the chain from field to cup — obviously in the hope of
high prices;
— a wish to bypass the existing distribution system and create a closer link between
growers and roasters, who would get access to excellent coffees otherwise
difficult to find.

Distribution Is Key

Auctions of this kind make it possible to trace rare, quality coffees directly to
farmers by conducting a competition and to inform roasters directly of the results. In
reality, however, the Brazilian auction could not have happened had the present
distribution system not been solidly in place — and left in place for the transactions!
Farmers and roasters are seldom equipped to handle intermediary export functions
such as transportation, letters of credit, payments, documentation, shipping and so
forth. For this reason, as part of the auction process, a respected exporter was nominated
to handle all ten coffees — and to accept the potential risks and gains from the
transactions. The idea that coffee can be sold by a small farmer directly to a small
roaster was not in fact tested. Skipping the traditional export-import portion of the
chain would not have been possible. So in the case of the auction, the coffee moved as
it does for ordinary sales: Farmer –> Exporter –> Importer –> Roaster.

122
Size Matters

To make a profit from coffee, one cannot deal with small lots; a certain volume is
necessary. This challenge becomes doubly difficult in a search for high-quality, gourmet
coffee. In this coffee auction, the lot size of the coffees posed a problem from the
beginning. Coffees are typically sold by the container, which is 250-300 bags. These
lots were on average just below 100 bags. This meant that an importer would either
have to piggyback his lot with other Brazilian coffees or pay the shipping on less than a
container. Several importers decided not to participate for this very reason. Internet
auctions of coffee are for exemplary quality only at the moment, and such quality
sometimes comes in very small packages. Future auctions will have to address this
problem and find co-operative shipping solutions fair to companies of all sizes. The best
coffee should not be overlooked simply because the logistics to get it to port are difficult.

Risks

In the coffee auction, all parties took financial risks, especially the farmers.
Without putting pressure on them and calling them every day, enough coffee would
not have come forth for an auction of this kind. The project asked farmers to take a
chance, to hold back selling their coffees in the hope that they would fetch better
prices. Shortly before the auction, they sold the coffee to the nominated exporter
under a formula worked out to split the premium. The opening prices were agreed
and fixed two weeks before the auction. This was not easy, as the benchmark prices on
the world market were heading up at the time.

Rigorous Planning Behind the Dream

The three-year Gourmet Coffee Project, under whose auspices this event took
place, was completed by mid-2000. Five very different coffee-producing countries
participated: Brazil, Burundi, Ethiopia, Papua New Guinea and Uganda. The project
tested a range of activities from production methods to marketing tools. The objective
was to try new methods of adding value to green coffee, enabling coffee producers to
generate premiums. The internet auction of Brazilian coffees was one of the many
activities of the project, and the lessons learned became available to all countries that
are members of the International Coffee Organisation (ICO). The project was also
presented to governments worldwide in an ITC presentation during its annual Joint
Advisory Group Meeting. The main sponsor of the project was the Common Fund for
Commodities. ITC was the executing agency.
Prior to the auction, the organisers (Table 6.1) worked out a long list (a very long
list, in fact) of all the things that could go wrong. In addition to traditional issues such
as disputes over the quality of the coffee, conditions for delivery and so forth, new areas

123
had to be considered. Were bidding procedures clear enough? What would be done if a
computer broke down or if the on-line connections got interrupted? How could one
secure full discretion for the bidders? Would closing with last bids be at night in the
United States, Europe or Japan? Many more questions of this type emerged as well.

Table 6.1. Organisers behind the First Internet Coffee Auction

Name Functions

Brazilian Specialty Coffee Association
Co-operative Regional de Cageicultores em
Guaxupe Ltda. – Cooxupe
The Gourmet Coffee Project
The Specialty Coffee Association of America
Liaison to farmers and the exporter; organisation of
the competition for selection of coffees.
Exporter of all ten coffees
Funding and advice.
Promotion via its web site; technical organisation of
the auction; contractual arrangements between all
parties.

What Did the Bidders See?

Using a password, bidders around the world opened a screen with the information
shown in Table 6.2.

Table 6.2. Information Provided On-Screen

Name Description

Item: A code for each of the ten coffees for sale.

Bags: Number of 60-kg bags for sale.
Bid: The highest bid per pound (lb.) expressed in dollars at any time. The opening
prices were different and had been agreed upon between the individual farmers
and the exporter.
Value: The system automatically calculated the total sales price at any point, i.e. bags x
bid (e.g. 49 bags at current bid of $1.55 cost $10 046.28).
High Bidder: The secret code of the bidder having made the highest bid at any point. The
bidders had been guaranteed full discretion during and after the auction, but all
bidders who purchased one or several coffees made it publicly known afterwards.
Description: Names of farms. By a “click” on the name, the bidder could obtain information
on the farm and the coffee (of which he had received a sample already).
Time: The auction opened with only one coffee (BGP 110) open for bidding. Every five
minutes, a new coffee was added until the screen was full. The auction closed 48
hours later — also at five-minute intervals.
Status: This column stayed Open for 48 hours. Thereafter it changed to Closed and the
closing price to be paid by the “winner” could be seen in columns 3 and 4.

124
 This screen shot, printed out during the auction (see Table 6.3), captures the
bidding process at one point, 17 December, at 17.30 Central European Time. The
screen shows that the auction had four minutes, 18 seconds left for bidding on Item
BGP 105; and 24 minutes, 18 seconds left on the last Item, BGP 101. It was morning
on the West Coast of the United States and the middle of the night in Japan.

Table 6.3. List of Available Auctions

Item Bags Bid Value (US$) High Bidder Description Time Status

BGP110 102 $1.85 24 960.29 C18E2A Carvalho Dias - Closed

BGP109 44 $1.65 9 603.16 968A94 Bittencourt - Closed
BGP108 134 $1.46 25 878.28 C18E2A Nagona - Closed
BGP107 150 $1.50 29 761.87 9C27BF Ferreira - Closed
BGP106 99 $1.38 18 071.41 D28CA3 Diamante - Closed
BGP105 49 $1.48 9 592.58 20F0F4 Fernandes 0:00:04:18 Open
BGP104 49 $1.32 8 555.54 C3E703 Watanabe 0:00:09:18 Open
BGP103 49 $1.55 10 046.28 20F0F4 Francischini 0:00:14:18 Open
BGP102 141 $1.80 33 571.39 F18EB1 Nannetti 0:00:19:18 Open
BGP101 50 $1.70 11 243.37 466DCC Ottoni 0:00:24:18 Open

Is the Coffee Auction a Viable Model?

The internet coffee auction was an experiment, made possible only on the basis
of great efforts and contributions by many parties in the form of risk-taking, man-
hours and flat-out sponsorship. People made these contributions with the obvious
hope that auctions of this kind could become viable in their own right in the long run.
The turnover at the specific auction was around $200 000 for 900 bags only — a
mere drop in the coffee ocean — but this was deliberate. If things had not worked
out, the cost of remedy and compensation would not have reached an alarming
magnitude. The auction attracted great attention in coffee circles, and the Brazilian
parties began preparing new internet auctions of their coffees from 2000. The volume
was to increase, the bidding procedures would be adapted, etc. It was assumed that a
third auction in 2001 could be self-sustaining for all parties involved. Indeed, in late
2001, the Brazilians arranged — for the third time — a similar internet coffee auction
but for larger quantities. All three Brazilian auctions held thus far have been successful
regarding both technology and prices obtained for the coffees. In mid-2001, Guatemala
organised an internet coffee auction of the same kind — also with very encouraging
results. Others who might consider organising an internet coffee auction now have the
valuable knowledge that it can indeed be done1.

125
 Nevertheless, the following cautions should be borne in mind:
— much tailor-made preparatory work is essential;
— parties will face risks bigger than those in the known day-to-day business, at
least at the start;
— the first auction probably will not be economically viable on its own and may
require external financial support;
— the coffees to be offered must be unique (in quality or as speciality or organic
coffees) to create attention and justify a premium; and
— lot sizes probably need to be large enough to justify separate shipment.

Note

1. Further information on the internet coffee auctions can be obtained from

www.intracen.org or scholer@intracen.org.

126
 PART THREE

COUNTRY EXPERIENCES
AND LOCAL EXPERIMENTS

127
128
 Chapter 7

The Micro-Foundations of E-Commerce:

Informational-Focused Development
in Andhra Pradesh, India
Kyle Eischen

They are called computers, simply because computation is the only significant
job that has so far been given to them. The name has somewhat obscured the
fact that they are capable of much greater generality…To describe its
potentialities, the computer needs a new name. Perhaps as good a name as
any is “information machine”.
Louis Ridenour, 1952.

Science, as well as technology, will in the near and farther future increasingly
turn from problems of intensity, substance, and energy, to problems of structure,
organisation, information, and control.
John von Neumann, 1949

“Software translates languages on Web”

Lotus Ireland, a subsidiary of International Business Machines Corp., said
Wednesday it had developed a new software product allowing people
speaking different languages to communicate across the Web. Lotus
Translation Services for Sametime translates Web chat and communications
as typed in real time, and will go on sale in September. The software employs
Java technologies.
San Jose Mercury “TechTicker”, 22 June 2000

As goes software...so goes business and, perhaps, even society itself.

Survey: Software, The Economist, 12 April 2001

129
The Micro-Foundations of E-Commerce

E-commerce has tremendous but unrealised potential to restructure the global

environment. Yet understanding if and how this potential will be realised, particularly
for developing economies, requires moving beyond e-commerce to consider the patterns
that underlie information technologies (IT) generally. These general patterns, what
makes IT a unique and transformative factor within the global environment, exist
separately from the actual development of e-commerce as a viable and extensive
economic and social reality. Mapping the central structures of IT that in turn structure
the nature of e-commerce is crucial exactly because it frames analysis around the
institutionalisation of new structures of inclusion, power and profit that operate
separately from the development of specific economic sectors or technologies.
An essential method for beginning to define these broader patterns is to signal
the fundamental difference between software and hardware.1. This seems simplistic,
but software is what underlies e-commerce, makes it dynamic and structures the nature
of economic relationships within it. Software also embodies the informational as
opposed to technological aspects of new information technologies. Informational
development, whether focused on specific sectors (B2C, B2B, ICs, OSs) or outcomes
(e-commerce, e-government, telemedicine), involves more than building hardware
infrastructure. It requires that relationships and social patterns evolve around and
become embedded in that infrastructure. From the protocols that offer interoperability
between computers, access to on-line databases, processing of transactions and real-
time communication to the “dancing paper clip” in Windows, software is a central
tool through which interactions with and between networks are structured. In this
way, e-commerce or other digital sectors, usually through software, are fundamentally
information-framed and socially structured processes.
More importantly, informational production, products and industries (like
software) reflect unique patterns that contrast with previous models of industrialisation2.
On a micro-foundational level, IT production follows patterns that reflect an
informational development process. This contrasts directly with industrial and
manufacturing patterns, even within other high-technology hardware sectors. It suggests
that informational industries entail different social, governmental and economic policies
and institutions in their establishment, fortification and expansion. Patterns from
previous industrial development efforts, particularly those focused on large-scale state
investment in fixed and defined resources, will conflict with these new patterns, limiting
the success of development efforts. State investment and policy remain central, but
must be aligned with the development of an overall informational structure. Focusing
on the technological or infrastructure aspects of IT, while comfortable within the
framework of industrial policy and institutions, will not result in successful new
economic and social development.
Bill Gates has argued that neither technology nor capitalism will alleviate the
problems of the world’s poor, but executives and policy makers from developing
nations have advocated exactly such an approach. Understanding the general trends

130
within IT helps evaluate such claims, to arrive at conclusions that reflect the complexity
and truth in both arguments. The forecast of IT’s transformation of India within five
years by the CEO of Wipro, one of India’s leading software firms, is probably accurate.
It does not necessarily follow that such a transformation will entail the alleviation of
the poverty Gates is concerned with, or that such a transformation will entail specific
technologies or organisational patterns like e-commerce as currently conceived3. Recent
statistics from India highlight such patterns (NASSCOM, 2001). In India as of
December 2000:
— there were only 1.8 million internet subscribers (and more than 5.5 million users);
— 68 towns and cities accounted for 92 per cent of internet users;
— the capital cities (New Delhi and other state capitals) accounted for 79 per cent
of internet connections across the country;
— there was a PC base of 5 million PCs. Of these, 3.7 million machines could be
effectively used for the Internet;
— 91 per cent of India’s corporate web sites were located overseas;
— the 1999-2000 total volume of e-commerce transactions in India was
approximately Rs. 450 crore (or roughly $96 000 000 as of December 2000
exchange rates), largely involving financial or B2B transactions;
— by 2005, forecasts predict that only 1.5 per cent of total e-commerce transactions
will come from B2C;
— by 2005, there will be a $9 billion business opportunity for Indian IT companies
serving the global e-solutions services market.
The statistics highlight that e-commerce as a generalised phenomenon for the
majority of people in India is unlikely anytime in the near future. However, e-commerce
will transform business in India, creating large new markets and fundamentally altering
the relationship between India and the global economy. How these new relationships
are structured and supported — organisationally, technically and politically — will
be the central way in which the majority of people in India are affected (positively or
negatively, through inclusion or exclusion) by e-commerce and new information
technologies.
As Dertouzos (2001) points out, digital technologies, as currently designed and
implemented, will not instigate revolutionary change until they are structured to meet
the needs of the 2 billion illiterate individuals in the world. This, again, does not
mean that these 2 billion individuals will not have their economies and societies
impacted upon by informational practices, products and industries. The very
conceptualisation and implementation of digital technologies as text and computer
based (assuming literacy, romanised character sets and a series of basic infrastructural
supports) is already exclusionary for a vast majority of the world’s population, including
large portions of Indian society (particularly women). It does mean that the structure

131
of that impact will be determined not by specific industries or products, but rather by
the institutionalisation of patterns that structure the production, manipulation and
dissemination of information as embedded in e-commerce and other applied technology
formats. As highlighted by the statistics above, these are the exact patterns that are
increasingly central issues for India in terms of e-commerce, IT and development in
the coming decade.
Such questions and contradictions become more open to analysis when the micro-
foundations of informational development, the mechanisms that translate information
into products and power, are considered. Such explanations are important exactly
because their absence limits an analysis of global processes generally, and the discussion
of specific policies and development strategies in particular. Moving beyond any
specific technological or sectoral focus to understand IT as an informational-driven
practice (Brown and Duguid, 2000) requires detailing the processes, products and
industry patterns that structure its development. This involves clarifying not only the
technological aspects but also the informational patterns within new technologies that
structure how information is created, distributed and managed.
A look at Andhra Pradesh, India provides an initial understanding of each of
these issues, helping clarify the space for action and the possibility of overall economic
and social growth within an “informational” development strategy. Although focused
on building a local software sector, the policies pursued by Andhra Pradesh outline
the overall social and economic impacts that informational strategies entail. Andhra
Pradesh demonstrates that sustainable and widespread IT-focused economic
development, even if targeted on one specific industry or sector, requires a
reconceptualisation of resources, capacities and potential outcomes in line with the
general patterns structuring informational development.

Opening the “Black Box” of Informational Processes, Products and Industry

The higher-level aspects of the information society are well established and
theorised for both macro analysis of how “information societies” establish a new
conception of social and economic reality and more detailed meso-level considerations
of how this reality is structured geographically and organisationally. In the conceptual
case, there is an extensive literature that seeks to understand the shift signalled by a
move from a physical to an informational-based environment. Such analysis focuses
on the nature of a “digital society” (Negroponte, 1995; Webster, 1995), the role of
knowledge as the central resource (Drucker, 1993) and the spatial and symbolic impacts
of such developments (Cairncross, 1997; Lash and Urry, 1994). This literature is
complemented by analysis of the institutional and geographic structures that transmit
and transform information on regional levels (Saxenian, 1994, 1999) and globally
(Gordon, 1994; Borrus, 1993) through new organisational forms and mechanisms
(Castells, 1996; Amin, 1994).

132
 While analytically rich, this broader theorising fails to link to changes in concrete
behaviour and practice. Defining the practices that underpin these general considerations
is essential to understanding what precisely is unique about an informational-driven
environment, in concrete terms with defined and predictable outcomes. As Rosenberg
(1982, p. vii) states, “(T)his is because the specific characteristics of certain technologies
have ramifications for economic phenomena that cannot be understood without a
close examination of these characteristics”. It is exactly these characteristics that are
missing in much of the analysis of “information economies and societies”.
Moving inside the “black box” of IT, it is possible to consider the “micro
foundations of economic dynamics” (Dosi, 1984, p. 1) that link technical change with
the macroeconomic and social forces that define a globalised information-driven
environment. Focusing on micro-level practices — and their embeddedness in norms,
organisations and practices that translate information into power and profit — opens
a path to a more rigorous and systematic understanding of how global trends and
micro-processes are interconnected, how IT is simultaneously technical and social,
and how “information societies” compare with previous industrial transformations.

Outlining Informational Production: Finding the Information in Information

Technologies

The Impact of Informational Patterns upon Norms, Organisations and

Institutions

How social practices and knowledge are codified into information technologies
becomes a central issue for understanding social and economic policies within an
informational environment. IT generally is not only a physical architecture that
facilitates global flows (Castells, 1996; Held et al., 1999), but also structures how
such flows occur. Assumptions about social processes are inherently built into the
algorithms that comprise information technologies. As IT increases in importance in
society, such assumptions increasingly impact upon how social relations are structured.
This in turn suggests that IT increasingly shapes the strategic and organisational choices
of governments, firms and civil society, whether consciously or not (Lessig, 1999;
Mitchell, 1998).
IT is increasingly the central method for producing, storing, transforming and
distributing knowledge. The structure of this process frames how social knowledge
becomes embodied in actual products. In other words, informational practices
increasingly define how information is produced and translated into power and profit
on the “shop floor” (or in the cubicle). Informational production comprises a very
tacit, unquantifiable process, which makes the regulation of both the inputs and outputs
of that process extremely challenging. In many ways, it has an artistic and “free-
speech” quality that challenges industrial forms of regulation and policy. In turn,
informational industries reflect the basic patterns of informational production and

133
products. The organisational patterns of firms and the global structuring of production
reflect patterns structured around the creation, manipulation and dissemination of
information. The vertically integrated, product-centred firm — linked first to global
networks of immigration and then to distinct regional centres of development
globally — make organisational and strategic sense when viewed through the demands
of information practices.
Informational production processes and organisational logics help elaborate on
many key patterns visible in a global, information-driven society. At the most simple
level, analysing informational practices explains how processes are both global (a
global market for IT products and labour) and extremely regional (a need for local
knowledge and interaction) simultaneously, and how such relationships are structured
through global networks and flows. Understanding informational production processes
helps link micro-level (shop floor) processes to broader organisational and institutional
trends in the global environment.
More importantly, this analysis helps elaborate on the informational aspects of
IT specifically and the “information economy” generally. The extensive literature
that considers Fordism and Taylorism indicative of essential social relationships,
production processes and organisational structures in the 20th century (Doray, 1988;
Giordano, 1992; Waring, 1991) establishes a pattern for linking “shop-floor” and larger
macroeconomic and social trends. Debates on the transformation of this structure,
linked specifically to “Japanese-style” production systems (Jurgens et al., 1993;
Sheldrake, 1996; Womack et al., 1991), indicate how such micro-level changes in the
organisation of production occur and impact upon broader industry and social patterns.
The predominance of IT in the economy, combined with its unique production process,
suggests that informational patterns play a similar role within an information-driven
environment.
Similar arguments have already been made for “high-technology” industries,
particularly the semiconductor industry. In arguing for an analysis of IT as a “new
mode of industrialisation”, Henderson (1989, p. 4) considers that “(S)emiconductors
are not only the heart of microelectronics and information industries generally, but
semiconductor companies themselves constitute a production and organisational form
that is a paradigm example for the global option in practice”. Angel (1994) continues
this line of reasoning by stating that semiconductor firms are optimised for innovation,
which in turn shapes the organisational and geographic structure of the industry. This
“structuring for innovation” is embedded in the linking of flexibility and innovation
on the factory floor, shaping formal and informal learning and information exchange.
While correct in focusing on the impact of IT, a focus on semiconductors or
other manufacturing-based high-tech industries fails adequately to frame analysis around
an “informational mode of development” (Castells, 1989) that operates within a
different supporting environment with distinct economic and social consequences.
Simply, building a semiconductor industry has much more in common with an industrial
model of development than with the promotion of software, biotechnology, multimedia,

134
networks or other informational industries. The difficulty regions have experienced
in moving from semiconductor to informational industries like biotechnology and
software, in contrast with the move from electronics manufacturing to semiconductors,
highlights this difference4.
Overall, an informational mode of development signals that regions develop
environments in support of informational practices and firms, as well as develop policy
to address the social implications of such an environment. Detailing the general
production, product and industry patterns that shape IT industries overall as
“informational” opens up the possibility of understanding the possibilities and challenges
of IT-focused development.

Informational Production: A Working Model

New information-based production patterns indicate how information is produced

in the economy, supported by organisational forms that frame both this production
and new social structures (and tensions). Innovation and competition will be centred
on products and not processes, where access to unique sources of knowledge and
human resources combined with market dominance are central to long-term competitive
advantage. These patterns will play out on both the global and regional levels as firms
seek global labour and product markets, but remain regionally focused, based on the
need for agglomerating production and gaining access to unique domain knowledge.
The widespread nature of these patterns, and their similarity to “networked
society” conceptions is not coincidence. An informational production model suggests
that such an environment will be global with strong regional agglomerations formed
by a need for skilled workers and domain knowledge. It will contain globally distributed
mechanisms for congealing social knowledge as both a resource and an end product.
It will include flexible and dynamic processes linked to the very tacit nature of its
production process. Importantly, it will consist of new organisations and institutions
to support such innovative, informational environments.
Focusing on processes, products and industries details the informational aspects
that structure and interact with the overall global environment. A working model of
“informational production” is comprised of several elements5:

Process

— It embodies a process organised around the definition, generation, manipulation

and transmission of information into socially and economically applicable forms.
— Because it is socially structured and often determined, the production of such
congealed knowledge will often take the form of craft-like (or creative or
research-like) production systems where tacit (as opposed to explicit) knowledge
is essential, peer or market review processes dominate and production processes
are non-rationalised.

135
— Skilled human resources from multiple knowledge domains will be the central
resource, with a weakly defined division of labour.
— Such production structures imply that extensive, as contrasted with intensive,
production will take precedence. This constraint will apply no matter which
organisational model, proprietary or open-source for example, is used to access
sufficient skill and resources to develop and test new information-based products.
— Increasingly, value added will be greater in the design or mapping of the
algorithmic aspects of a process, i.e. in the ability to define and model it, than
in its actual implementation, manufacture or replication.

Product

— Informational products will have both functional and expressive qualities that
will increasingly blur and extend existing legal codes and norms based on
industrial models of intellectual property.
— Central debates will focus on balancing privacy, intellectual property, public
and private goods and speech concerns around IT products.
— The social knowledge and assumptions embedded in IT products will result in
the institutionalisation of norms prior to the application of the technologies in
society.
— Business models, reflecting the difficulty of controlling distribution and
replication of intellectual property, will move to establish products as de facto
market standards and then control the pace of innovation with such standards.
— Given the limits of process rationalisation, products will be the central source of
competition.
— Competition will centre on whether the translation and codification of existing
processes into digital or algorithmic forms constitutes innovation or mere
replication of existing knowledge.
— Products will inherently contain flaws derived from their non-rational production,
the complexity of processes being mapped and the assumptions built into their
design.
— The nature of competition will tend to produce dominant products and standards
that structure the pace and direction of multiple industries.

Industry

— Flexible, networked organisational forms able to manage flows of information

efficiently and rapidly (both its creation and applications), guide “information”
or “knowledge” workers and implement new tools for manipulation and
dissemination of information will take precedence.

136
— Organisations will locate in regional environments that both produce knowledge
and stimulate its transformation into congealed, marketable forms.
— Regions globally networked through flows of individuals will tend to predominate,
individuals being the key social institution for transmitting tacit knowledge as
well as providing the highest value added within the global environment.
— Regional centres of information development (either independent or within a global
firm) will be responsible for the full development cycle of individual products.
— Labour markets will increasingly be globally defined, although product markets
will be fragmented.
— The social nature of domain knowledge, i.e. the place and context-specific
understanding of processes, ensure that regions and culture will play a significant
role in the development of IT.
— Firms will tend to be vertically integrated and dominant in specific knowledge
domains.
— Monopolies will tend to occur, both through the establishment of standards and
the drive of firms to control the innovation and product cycle.
Analysis of globalisation and the information society (Castells, 1996;
Gordon, 1994; Held et al., 1999; Amin, 1994; Drucker, 1993; Shapiro and
Varian, 1999) has already established macro patterns similar to this working model
for the nature of innovation and knowledge, patterns of global flows, network
organisational forms and new social structures. Outlining informational patterns,
however, serves as a clarification of why information is valuable and how it is structured
and valorised in the global economy on a micro-level. It also opens up a method for
linking these broader theories with individual case studies that lead to a clearer definition
of the strategies and outcomes at play for new development efforts within the global
economy, like that of Andhra Pradesh detailed below.

Building an “Information Economy”: The Case of Andhra Pradesh, India6

On the southeastern coast of India, the state of Andhra Pradesh is undertaking a

dramatic experiment in building what can justifiably be called an information society7.
Over the last six years the state government has propelled Andhra Pradesh from virtual
anonymity to status as a central location for software investment and development
within the global economy. By 1999 Andhra Pradesh had 15 000 software professionals
in 192 firms, with the number of companies growing by 70 per cent a year8. A central
aspect of the regional initiative has been the social transformations linked to fostering
an overall “informational” environment in support of the IT strategy. Yet, for most of
the 1990s, no one would have foreseen the possibilities of development in Andhra
Pradesh in general, let alone development centred on one of the most dynamic and
leading industries of the coming century.

137
 The central questions surrounding the development strategy focus on the ability
of the region to promote IT development within exactly this economic and social
environment. The resources and capacities that the regional strategy built on and created
signal how the overall informational patterns above impacted upon policy and
development efforts. Andhra Pradesh also directly addresses the question of IT generally,
and software particularly, as an overall engine of regional growth. Most important,
the nature of the strategy and the corresponding policies — regarding which industries,
investments and programmes have or have not been targeted — frames informational
strategies within a context that outlines the specific opportunities and challenges of IT
for developing economies globally.

Initiating and Sustaining Informational Strategies

The general features of “information regions” or “technopoles” that support

knowledge or information-focused development initiatives have been well-established
(Castells and Hall, 1994; De Vol, 1999, 2000; Mytelka, 1999). In general, IT development
strategies strive for three simultaneous goals: re-industrialisation as linked to a constant
transforming of comparative advantage, regional development to ensure equitable and
broad growth and the building of synergy to create new economic opportunities and
vitality (Castells and Hall, 1994). In practice these goals evolve within the categories of
public policy, comparative location benchmarking and social infrastructure development.
The central catalyst in each is the state. “State and local governments, public policies,
and the interaction between private and public sectors are crucial for the genesis,
expansion and fortification phases of high-tech development” (De Vol, 2000, p. 3).
Distilling these general factors into a simple matrix helps simplify the complex
milieu of factors that support informational regions and highlights the features, if
any, that can be carried over from pre-existing development strategies. Very broadly,
governments should focus on five factors central to the establishment and growth of
an IT region:
— research centres;
— investments to jump-start development;
— human-capital development;
— capital and institutions for the commercialisation of research; and
— a governing and institutional culture and social environment that promote
development.
Hidden within each of these factors is a double quantitative and qualitative
characteristic crucial to evaluating the viability of the “genesis, expansion and
fortification” of IT initiatives. The quantitative aspect involves fairly direct
infrastructure and institutional initiatives that can be measured by standard financial
and statistical measures. In Andhra Pradesh, these quantitative aspects are exactly

138
those supplied by the national legacies that were crucial to the genesis of the regional
strategy (Eischen, 2000b). The qualitative aspect involves the nature and quality of
the institutions, cultures and relationships elaborated around the development initiative
that ensure its expansion and fortification. Successful IT strategies involve the state as
a catalyst of both economic and social transformation at different stages of the initiative.
This implies that information-focused initiatives involve the region-state exactly because
of the qualitative factors (social infrastructure, institutional culture, public-private
partnerships) surrounding the institutionalisation and growth of the regional IT strategy.
As for most development initiatives, the initial kernel (or genesis) of an IT
regional strategy may be quantitatively measured, but long-term success lies in the
qualitative aspects of research, investment, human capital, commercial networks and
governance structures. IT regions are inherently social projects for building new social
norms and institutions to ensure long-term comparative advantage. Understanding
and evaluating the Andhra Pradesh initiative involves analysing how regional efforts
have framed both these quantitative and qualitative aspects within the overall patterns
of informational development outlined above.
Information industries such as software require supporting social institutions
and organisations to link local resources with the global economy. These institutional
linkages are a mechanism through which IT industries access unique local social
knowledge on which global competition centres. The specific features of IT processes,
particularly the need for locally specific domain knowledge, represent one of the
essential mechanisms through which local information is valorised in global economic
and social circuits (Geertz, 1983). Information technologies such as software thus are
incredibly flexible and dynamic processes, not only comprising how information moves
in the global environment but also embedding local knowledge in the technologies
themselves. Information industries build on pre-existing institutions, cultural patterns
and social norms to access unique domain knowledge, while simultaneously creating
new structures that slowly intertwine and transform these existing social relationships.
Fully understanding the implications of these broader patterns requires detailing
the opportunities and challenges presented by the shift to information-focused processes,
products, norms and institutions. The choice of software as a central focus of
government policy and economic growth is not just a simple choice of policy or
industry, but involves a restructuring of social and economic relationships in line with
the qualitative aspects of the factors that support informational industries generally.
Such an understanding is clear when considering the three facets of the regional
initiative: the replication of national resources on the regional level, the building of
unique regional capacities and the fostering and capturing of private initiatives. The
key thread running through each of these, given the essential quantitative and qualitative
aspects of IT regions, is that they form an organic whole essential to developing the
long-term viability of the information-focused strategy in Andhra Pradesh. This long-
term viability goes beyond the mere fostering of “information industries or
infrastructure” and entails the reformulation of social and political relations around
an innovative IT environment — explicitly building the institutional framework that

139
supports an informational region operating in a global economy. In this way, Andhra
Pradesh is an ideal case in which to consider the social and economic framework
within which new social relations are structured as regions pursue new informational
models of development.

Extending, Incorporating and Building National Legacies into New

Institutions

Given the challenges posed by the inherited national legacies and informational
production structures, the specific policy initiatives of Andhra Pradesh (Table 7.1)
outline the simultaneous building of an innovative environment and addressing the
potential limitations of software as a driver of regional development. The uniqueness
of the Andhra Pradesh initiative rests on the local resources, capacities and institutions
that have spurred regional development relative to other regions nationally9 and
globally10. This regional advantage is supported by a real understanding of the innovative
environment, i.e. the qualitative aspects of IT regions, that informational regions
require. This overall understanding has allowed regional policymakers to incorporate
national legacies, build innovative regional programmes and tap into global networks
to both initiate and sustain the regional initiative.

Table 7.1. Overview of Andhra Pradesh Regional IT Strategy

Objectives Initiatives

Capture National Resources ½ Software Technology Parks, Hyderabad

Set Up Unique Programs and
Institutions
Foster Private Initiatives and
Networks
½ HI-TEC City Development
½ Indian Institute of Internet Technology
½ Indian School of Business
½ Institute of DNA Finger Printing
½ Government Bureaucrats and Administrators
½ AP Biotechnology Park
½ Andhra Pradesh Technology Promotion and Development Centre
½ Apparel Export Park
½ Regional-State Venture Capital
½ Regional “SMART” Initiatives
½ The “Talking Heads”
½ Satyam Corporation
½ ICICI Knowledge Park
½ ALEAP Industrial Estate

Overall, the Andhra Pradesh regional strategy focuses on the development of a

“hi-tech habitat” that develops the infrastructure, partnerships, regulation, financing
and social norms that will create an innovative region over the next two decades
(Government of Andhra Pradesh, 2000). Key investments seek to build the infrastructure
that will be the architecture within which software and informational development
generally can take place. New partnerships with national, regional, private and

140
international institutions for both expertise and financing support such infrastructure
investments. The partnerships and initiatives have been directly supported by regulation
within the control space delineated under the Indian constitution (Eischen, 2000b).
These partnerships have also involved capturing national institutes to operate within
regional objectives, either incorporating them into regional development coalitions or
directly framing their policies to match regional initiative aims. The hope is that in
combination these initiatives will create a synergy that will foster an innovative
environment in which talented individuals will stay in the region and overseas Andhras
will return and establish new ventures.
The overall direction of the regional strategy is important, but the new structures
created in support of the initiative clearly highlight the social transformations entailed
by the strategy. The combination of national legacies, regional innovations and private
initiatives details the building of an information region in practice from the ground
up. Overall, Andhra Pradesh provides detailed evidence of the sustainability and reach
of the software-focused strategy as it moves from genesis to expansion.

Replicating and Capturing National Institutions and Resources: Filling in

the Gaps

Much of the effort in Andhra Pradesh over the last six years has focused on
replicating key national institutions or capturing existing national capacities for regional
goals. While national legacies did provide key resources to initiate development in
Andhra Pradesh, there were gaps in this inheritance compared with other regions
nationally (Eischen, 2000b). Much effort has focused on replicating national institutions
and capacities that facilitated IT industries in other regions in India. The national
institutions and resources that did exist in the region have essentially been captured
and transformed into de facto if not de jure regional resources. Key developments
along both these lines have been:

Software Technology Parks, Hyderabad

The STP in Andhra Pradesh has become an integral part of the regional initiative,
promoting software investment in Andhra Pradesh in direct competition with
other regional STP centres in India. It is a key actor in the HI-TEC City project
and has become a lead partner with the state in expanding STP services
(telecommunications, single-window clearance) to new urban regions throughout
Andhra Pradesh hoping to capture investments and small local software start-
ups funded through new state-sponsored venture capital.

HI-TEC (Hyderabad Information Technology and Engineering Consultancy) City

The HI-TEC City is the premier targeted regional investment focused on
constructing a software cluster. The $850 million software park, explicitly
modelled on initiatives such as the Electronics City in Bangalore, has been central

141
to attracting new investors, including Microsoft, Metamore and Oracle.
Additionally, companies such as Motorola, Wipro and Baan Infosys have
undertaken new or expanded investments in Hyderabad and are building campuses
near the HI-TEC City. Structured as a land-for-equity swap (the regional
government provided 158 acres of land in return for an 11 per cent equity stake),
the venture is a public-private partnership between the Andhra Pradesh Investment
Corporation (APIC) and L&T Engineering. Overall planning has focused on
using urban land policy to link geographically new investments in software
campuses, leading universities and research centres, new housing initiatives and
telecommunication and transportation infrastructure, around which the software
industry will expand region-wide.

Indian Institute of Internet Technology (IIIT)

The IIIT is an investment paired with the HI-TEC City. The aim is to build from
the ground up an institution equivalent to the six existing Indian Institutes of
Technology and Science found in other regions throughout India. The IIIT has
drawn investments from many of the same firms investing in the HI-TEC City.
Specific investments include: the Microsoft School of Software Technology, the
Oracle School of Advanced Software Technology, the Metamore School of
Excellence in Software Methodologies and the IBM School of Enterprise Wide
Computing.

Indian School of Business (ISB)

The Indian School of Business is being constructed next door to the IIIT as a
local variant of the five Indian Institutes of Management. The school is the idea
of Rajat Gupta, a managing director of McKinsey & Co. The $80 million school
was moved to Andhra Pradesh after strong regional advocacy by Andhra
government officials and in reaction to resistance in Mumbai to the absence of
student enrolment guarantees for scheduled castes. The school is being formed
in partnership with the Wharton School of Business at the University of
Pennsylvania and the Kellogg Graduate School of Management at Northwestern.

The Institute of DNA Finger Printing

This is a new nationally funded research institute specialising in basic R&D in

DNA fingerprinting, molecular diagnostics and bio-informatics. It serves as the
bio-informatics national node for the European Molecular Biology network
(EMBnet), as well as the national centre for bio-informatics services including
the biomolecular sequence, macromolecular structure and genome databases. It
also contains a new graduate school in all fields of the life sciences. The institute
fits perfectly with the regional focus on developing information-based industries
backed by institutions to generate new research and skilled human resources.

142
 Management Skills to Run Government Programmes
One of the truly extraordinary aspects of the development initiative in Andhra
Pradesh has been its formulation, implementation and management by career
national bureaucrats and locally based national research institutes. Capturing
these national skills and resources and directing them to local needs has been
one of the key factors in establishing the management and technical skills to
develop and implement the regional initiatives.

Building Regional Institutions and Capacity

Adaptation of existing resources and building local models to fill in gaps in

the national inheritance have combined with unique regional initiatives. Long-term
success of the software strategy depends on building regional innovative capacity
around information-focused production. Regional policies suggest that applying such
skills to economic and social issues is a crucial aspect of developing innovative capacity.
One of the central features of the regional initiatives is global benchmarking. Regional
planners have freely borrowed or adapted programmes from successful initiatives in
India, Southeast Asia, Europe and the United States. They have also created unique
programmes to serve as models for other regional initiatives. Key programmes include:

AP Biotechnology Park
Like the HI-TEC City, the park is designed as a focus for commercialisation of
biotechnology research and manufacturing as well a basis for the application of
biotechnology to agricultural needs within the state. The park is set up explicitly
to generate “synergy with other technology-based companies and public research
organisations, creating opportunities for networking, collaborations and
technology exchange”.

Andhra Pradesh Technology Promotion and Development Centre

This government project provides a central location to help local industries and
entrepreneurs reach and compete in global markets through technology innovation
and the meeting of international standards. It is modelled directly on the Steinbeis
Foundation for Economic Promotion in Baden Wurttemberg, Germany, designed
to help SMEs in technology acquisition and global sales. The Centre operates as
the central node in a network of dozens of regional and national research institutes
and firms.

Apparel Export Park

The park is focused on the expansion of textile exports from the region. It is
directly linked to the National Institute of Fashion Technology and the Apparel
Training and Design Centre, built next to the HI-TEC City, that offer new
technologies, design and human resources to the Apparel Park.

143
Regional-State Venture Capital

The government has created local VC funding totalling Rs. 15 Crore

(approximately $4 million at current exchange rates)11 that will be privately
managed. This has been supplemented by regulations that will allow banks to
invest in VC funds and tax pass-through benefits to avoid double taxation,
irrespective of the form of venture capital.

Regional “SMART” Initiatives

The regional government philosophy is framed around the concept that “IT is
SMART”, meaning that IT is a political necessity to make government Simple,
Moral, Accountable, Responsive and Transparent. The aim is to spread the benefits
of the IT initiative to the population as a whole through governance, services
and quality of life improvements. This combination of national skills, local
domain knowledge and global best-practices, focused on applying IT to real
social problems, is one of the central factors separating Andhra Pradesh from
other regional IT efforts. Through the SMART initiatives, the government has
reconceptualised information-focused development beyond economic goals
towards a vision of IT as an integral part of governance and society. Key
programmes include:
— Chief Minister’s Information System (www.andhrapradesh.com) — an on-
line database providing information on electricity, water, health and finance
among other things, to gauge the daily progress of various infrastructure
projects and government services on a state-wide basis.
— APSWAN Fiber Network — the backbone of an information and video
conferencing network used for daily conferences between the central
government and the 25 regional districts in the state, focusing on local issues
and problems. The state is also experimenting with video-conferenced criminal
trials in order to maximise scarce judicial resources.
— The Computer-Aided Administration Department (CARD) programme has
streamlined the granting of land titles and paying land taxes on a state-wide
basis. These various processes, inherited from the British and still performed
by hand in many parts of India, had previously taken months to complete
but now are finished while citizens wait.
— The Twin Cities Integrated Services (TWINS) programme seeks to build IT
centres throughout the state, providing 18 services to the public. Birth and
death certificates, drivers’ licences and welfare cards are produced while
citizens wait at the initial pilot centre in Hyderabad. The system also allows
for the payment of basic utility bills.
— Rural Internet Community Centres, set up under an agreement with WorldTel,
provide internet access in rural areas in the region that provides real-time
and accurate prices to farmers, eliminating traditional brokers.

144
 — Secretariat Knowledge and Information Management System (SKIMS) — a
centralised database system that seeks to manage efficiently and integrate the
information and knowledge of each government bureaucracy in order to
improve efficiency and transparency.
— Janmabhoomi Programme — a rural initiative designed to combine financial
and materials support with local community management and labour to
develop educational, transportation and basic services infrastructure.

Private Initiatives and Networks: Capturing Synergy

In addition to these government programmes, private initiatives have supplemented

regional initiatives and in some cases provided models for them. They are:

The “Talking Heads”

This is a local network of CEOs established by returning Indian executives to
model the process of dialogue and networking that occurs in other regional IT
centres like Silicon Valley, Austin and Seattle. They serve as an unofficial advisory
board for regional policies and an entry point into global corporate, financial
and human networks.

Satyam Corporation
Satyam, as one of the leading software firms in India, has provided an indigenous
regional model of global-level software development. The company has built
an entire infrastructure, in essence a corporate city (with employee housing,
golf course, direct satellite uplinks and helicopter pad) that is a model of IT-
based business within the region and the country. This is combined with a business
culture that is distinctly local and global simultaneously. Satyam was the second
Indian software firm listed on the NASDAQ, but retains its global headquarters
an hour and a half outside the capital city of Hyderabad. It aims to be a global
company, yet has been an active and even leading force in the development of a
regional software industry. When the region was seeking Microsoft’s investment
in the HI-TEC City, Bill Gates was taken directly to the Satyam campus as a
direct example of regional software development potential.

ICICI Knowledge Park

This is a private R&D park and network developed by the International Credit and
Investment Corporation of India, the only universal Indian bank, and designed to
facilitate research and technology exchange between universities, research institutes
and the private sector. It has links with all the major national institutes in Andhra
Pradesh, all of the IITs and IIS, the International Centre for Genetic Engineering
& Biotechnology, Delhi and the National Institute of Immunology, Delhi.

145
 ALEAP Industrial Estate

This is the country’s first industrial estate exclusively for women entrepreneurs,
developed by the Association of Lady Entrepreneurs of Andhra Pradesh (ALEAP)
as a non-profit venture with support from the national government. It has served
as a model for APIIC, the state agency charged with developing other industrial
parks in the region.

Policy Implications: Understanding the Dichotomies and Possibilities of

Informational Development

The story of Andhra Pradesh opens a window through which to consider how
informational development strategies are initiated and sustained and how they impact
upon regional structures over the long term. Underlying these aspects are potential
dichotomies essential to evaluating the ability of information strategies to generate
widespread economic and social development. For initiating a regional IT strategy,
the dichotomy revolves around the central role of national legacies both in providing
resources and in the inherent limits and challenges included in such legacies. The
Andhra Pradesh experience clearly highlights this tension. National legacies played a
key role in establishing the human, research, legal and infrastructure resources that
the region drew on to establish a credible and viable software cluster. Yet they also
limited the choices, given the failure of national industrial policy, and left dramatic
social inequalities and challenges for the region to address.
For the choice of developing a software cluster, the dichotomy is that software,
because of its unique informational production and organisational patterns, both fails
to address social inequalities directly and is the region’s best hope for overall economic
development in the coming decades. The structure of a software cluster builds on or
even exaggerates the social inequalities inherited from the national development model.
Yet software, in its increasing dominance as a central global industry, its information-
based production process, its ability to absorb and valorise local knowledge and its
relatively unique economics, also offers an excellent chance for long-term regional
growth relative to other industries.
Developing sustainable regional initiatives that can recognise and address these
dichotomies requires understanding software (or informational industries more broadly)
as not a product or an industry, but a process that epitomises knowledge valorisation
within a global, networked, innovation-driven environment. The implication for
regional development thus revolves around: i) creating regional innovation systems
that support the constant evolution of products, services and firms (Malecki and
Oinas, 1999); and ii) understanding the process by which such regional innovation
systems network multiple local and global economic and social circuits. It is possible
to work through these dichotomies by considering how Andhra Pradesh has managed
exactly these issues.

146
 Developing Regional IT Initiatives: National Legacies and Regional
Innovation

Andhra Pradesh has not only established a software industry in a state that is
70 per cent rural, 40 per cent illiterate and was bankrupt in 1995, but it also has
clearly pointed out key factors for other regions seeking to promote information-
focused development. One of the central ironies of the Andhra Pradesh initiative is
that policies and institutions that were weaknesses under national or industrial
conceptions of development — chiefly skewed educational systems, lack of basic
infrastructure, skewed legislative initiatives and regional factionalism — have become
strengths under the regional software-focused development model. These are mirrored
by positive national legacies including military research institutions, targeted
telecommunications infrastructure, world-class universities and engineering and
scientific human resources that provided the essential factors initiating the strategy
(Eischen, 2000b).
The importance of Andhra Pradesh, however, is that policies undertaken over
the last six years clearly indicate that national legacies alone are not sufficient to
generate long-term development or address the shortcomings of failed national strategies.
Regional initiatives are crucial to building and sustaining regional development,
especially when focused on building regional innovation capacity that will support
the expansion of an informational strategy. Andhra Pradesh has targeted specific
investments in infrastructure, telecommunications and education that institutionalise
and expand on many of the most crucial national legacies. The investments that Andhra
Pradesh has pursued, as exemplified by the Microsoft investment, have focused on
capturing not only leading firms but leading products that form central technologies
of the future, such as Windows NT, Motorola’s Digital Signal Processing designs or
Oracle’s Enterprise Resource Planning software. Furthermore, the strategy has focused
not only on attracting investment, but also on linking business investment to additional
investments in education and research. The creation of the IIIT and ISB are essential
programmes that ensure the long-term commitment of firms to the region as well as
the capacity to develop the innovative and human-resource bases that will insure the
state’s place in the global economy. These investments are complex alliances between
the state, firms, local and international universities and non-resident Indians, all guiding
these institutions to become spaces for global best practice and innovation.

Software as an Engine of Regional Growth: Addressing the Limits and

Possibilities

The software-focused strategy creates a strong possibility that by building upon

and adapting the positive aspects of the national legacies existing social inequalities
will not be addressed. Even more significant, inequalities could become institutionalised
within software-based development as resources and policies focus on expanding those
legacies (e.g. investments in higher education and not literacy) essential to an innovating

147
region. Under these conditions, there is serious potential for a permanent digital divide
(because access to technology, education and income are highly correlated), not between
nations or regions but between social classes, cities or even neighbourhoods, as
inequalities in education, quality of life, basic services and infrastructure are reinforced.
The overall social transformations linked to industrialisation generally — the
rise of a middle class, urbanisation, consumerism, increased gender equity, expanded
literacy and social movements — that have accompanied high-technology hardware
industrialisation (Lubeck and Eischen, 2000; Rasiah and Best, 2000) have yet to be
established as the default pattern for IT-focused strategies. In contrast with previous
hardware-centred strategies, no clear trend or method has demonstrated the capacities
and policies states should develop to manage these challenges (US Department of
Commerce, 2000; Lessig, 1999). The formation of new IT clusters globally has
replicated the labour and firm agglomeration patterns that have characterised the
industry’s development in places such as Silicon Valley (Benner, 2000). The most
established regional centres of software development — chiefly in Ireland, Israel and
Karnataka — have come to replicate the social and economic tensions present in Silicon
Valley12, starting from a significantly lower level of development and with much
weaker mechanisms for social equity.
Strong and socially consensual policies are needed to embed development in the
broader society rather than merely set up a parallel social and economic reality. If
such policies do not exist, the risk is that social and economic development will remain
relatively stagnant, while new institutions and targeted investments link small sections
of the society to the global economy. In Andhra Pradesh, it is unlikely that export-
focused and FDI-driven software can generate sufficient growth to address all of the
state’s challenges. Some 15 000 software professionals and $70 million in exports are
not sufficient to address the challenges of inequality in a population of 80 million
with 17 million living below the state poverty line. The ability of governments to
generate policies that distribute and maximise the gains from information-focused
strategies while maintaining the organisational and social structures that provide those
gains, is one of the greatest challenges to regional IT initiatives.

The Hidden Possibilities in Information-Based Development Strategies

There are possibilities inherent in and unique to informational industries such as

software that can address this possible institutionalisation of inequality. As outlined
above, the opportunity around software, and industries following IT production patterns
generally, is that local knowledge is needed and valorised within the production process
itself. Regions should recognise that they are really developing a mastery of
information-based production processes, not specific IT industries. In other words,
focusing on synergy and developing the basic structures that promote an “ecology of
innovation” in Andhra Pradesh will produce economic and social development far
outside the software sector.

148
 Software is and will remain a central factor within IT globally, much as it
underlies the development of e-commerce. As outlined above, however, the blending
of software with other industries and domain knowledge is what produces the real
dynamism within information technologies. This, combined with the need for local
domain knowledge, opens up the real possibilities for regional development strategies
to overcome the potential limits of an IT strategy. Existing industries and research
centres, from media and film to agriculture and education, can become an integral
part of a successful strategy, expanding the impact of the development process and
ensuring its sustainability over time. All of these factors combined will embed local
knowledge into broader national and global processes, while giving some possibility
that the most negative social and economic divisions are ameliorated.
The essential point for policy, one that Andhra Pradesh epitomises, is that within
informational development strategies the space for building the innovative capacities
of the region and addressing social inequality can and should be linked. Building an
innovative environment means creating synergy between various sectors of the economy
and society that can generate both the resources and ideas that produce new innovation.
Adapting software and IT to local needs provides exactly such a synergistic environment
in which local knowledge is valorised on a global level and global skills are applied to
local needs. This is the unique opportunity provided by an information-focused strategy.
Defining and clarifying the nature of IT processes like software help explain why
information is valuable and how it is structured and valorised in the global economy
on a micro-level. This process of commercialisation (or transformation into authority)
of information creates strong similarities in production processes across software,
biogenetics, film, music, academia, media, design, architecture and finance. The Andhra
Pradesh software strategy, properly conceived as an informational strategy, provides a
basis from which to provide widespread regional growth.

Lessons for Regional IT Development: Legacies, Adaptation and New

Informational Industries

In evaluating the possibilities for sustained development, all of the basic factors
outlined in the matrix are essential. Clearly a region’s national inheritance, both positive
and negative, shapes the initial genesis of an informational strategy. The long-term
viability of the initiative depends on the local adaptation of these resources to address
both the challenges of the legacies and the impact of the IT strategy itself. Addressing
both of these challenges means linking different information-focused industries, skills
and markets to revitalise existing industries and develop new growth opportunities.
Institutionalising such synergy should be the goal of the overall development strategy.
Such institutionalisation and the capturing of sectoral convergence involve social as
well as economic transformations. In this way, viable IT strategies create “information
societies”, not information economies.

149
 Andhra Pradesh highlights how various e-governance, education, infrastructure
and investment policies overlap to build new markets, address social inequality and
build a new social contract around an information-focused strategy. Its specific policy
initiatives demonstrate that regional planners do understand the dichotomies and
opportunities considered above. They have both built on and adapted national legacies
to ensure that the initiative reproduces and expands the initial educational, research,
infrastructure and management capacities that underlined the genesis of the strategy.
The initiative has consciously sought to develop synergy between various types of
information-based industries (specifically biotechnology and finance) or embed
information technologies in traditional industrial sectors (apparel and agriculture). In
the process, it has dynamically reinforced the overall innovative capacity of the region
and widened its impact.
In many ways the initiative has sought directly to build an “information region”
incorporating the general patterns that structure information processes and organisations
into all social, economic and political facets of the region. In this way, it has built
institutional bridges between global trends and local needs. A central example lies in
the e-governance initiatives. These projects, by using local skills and knowledge, will
gradually expand the market for local IT products and services as the government and
society expand their capacities and demand for IT. A small example is the running of
the entire computer system for the IIIT on Linux, managed by local university
graduates. This both promotes Andhra Pradesh as a regional centre of software skill
and offers the ability to apply such skills to local institutions and needs. Another
example is the aim to develop a globally competitive apparel industry reinforced by
the use of local IT skills, professional design, unique national and regional cultures
and IT infrastructure. Global information technologies and architectures can also
potentially feed directly into the thriving regional Telugu film industry that serves
80 million people locally as well as the global Telugu diaspora.
In each of these cases, the successful regional software strategy links regional
government, culture and industry with global trends that reinforce and feed off regional
initiatives. This potential is inherent in informational production processes generally
(Table 7.2). In Andhra Pradesh, the failure to harness this potential would weaken the
long-term viability of the regional software initiative, while adapting this potential to
regional needs is crucial to establishing both a globally competitive software industry
and software as a catalyst for broader regional growth. Overall, the development
strategy in Andhra Pradesh is an ideal example of understanding an information economy
and adapting its potential to regional political and social issues. While the long-term
outcome of the initiative is not assured, Andhra Pradesh details the essential role of
regional governments in building social, economic and institutional synergy around
information-focused regional development. It is exactly such qualitative aspects that
will most likely be the central determinant of successful regional initiatives over the
coming decades.

150

General Patterns
Informational Processes,
Products and Industries
Institutional Mechanisms, Norms
and Flows Structuring the
Informational Environment
Local Processes and Structures
Interacting with the Informational
Environment
151
Table 7.2. Global Patterns Operating in Andhra Pradesh
Specific Aspects
½ Tacit, innovation-driven and informational
processes.
½ Products embedded and defined by social
knowledge.
½ Global markets and regional, networked
production for industries
½ Immigration
½ Investment
½ Innovation networks
½ National and regional policy
½ Private-public partnerships
½ Entrepreneurial firms-global firm linkages
½ Regional economic and social networks
½ Unique local knowledge information
½ Regional governance institutions and capacity
½ ½
Unique social capital
½ Unique cultural practices
½ Regional educational and scientific institutions
Specific Form in Andhra Pradesh
½ Development of software using skilled researchers,
engineers and firms
½ Use of AP exposure to foreign cultures.
½ Global software industry linked with local firms and
institutions to access and market local labour
½ Skilled global immigrant networks
½ Ties to US venture capital
½ FDI by leading IT firms
½ National & regional development initiatives
½ Regional government-Satyam Partnership
½ Global firm outsourcing linkages to local firms
½ Regional Andhra identity
½ Weak existing industrial structure
½ IT used for better governance and regional development
Regional political party
½ Extensive immigrant networks in software firms globally
½ Over production of skilled SW labour and English-language
university education
Implications of Informational Patterns: Mapping Interactions and Structures
in a Networked Society

The four quotes that open this analysis offer entry points into a fuller understanding
of information technologies and their impact on social and economic development.
Ridenour describes the process of how computers can be and, as detailed above, have
become “information machines”. The key technological development of the last 30
years is exactly the creation of information technology as a general phenomenon with
ever-greater distribution and connectivity. The key aspect is not, as is the usual focus,
technology, but rather information. The increasingly connected and digitalised world
we inhabit is structured by the flow of information and by how that information, as
socially determined, gets congealed into new forms of profit and power.
Von Neumann, speaking over 50 years ago of science and technology, helps to
explain the basic patterns of economic and social development as issues of structure,
organisation, information and control under these informational flows. An information
economy restructures social relationships in order to maximise the generation,
manipulation, dissemination and commercialisation of information. This results in
new organisational and institutional imperatives that offer both opportunities and
challenges. A distinct informational pattern of production models how knowledge
and information are produced and valorised in the global environment.
The IBM press release describes this entire process, in essence what the information
economy is, in three short sentences. An Irish subsidiary of IBM has developed a web-
based, real-time translation program based on Java. The knowledge to create (or
recreate) such a program resides in the programmers who created it, in Ireland, even
though they work for a global firm. The product is entirely digitally based and functions
only on the internet, where it is instantly available globally. It enables real-time
communication regardless of location or language. What these languages are, how
well they are translated and how they structure communication is a function of the
domain knowledge and assumptions of the programmers as structured in the program
code. The ability to regulate the product is inherent in the coding of the program and
the priorities of IBM, not in any government or social process.
The last quote serves to emphasise the social and informational nature of
information technologies. Kranzberg (1985) states that “technology is neither good,
nor bad nor neutral,” meaning that society determines its application and impact. Yet
informational technologies like the software produced by IBM already define
frameworks for social action through their architectures and development processes.
Informational products like software or gene patents are the vehicles through which
knowledge and information are given form in a digital world, and as such they explicitly
embody social knowledge. This means that prior to its application by society, IT is
already structuring the forms that social interactions can take. Combined with Metcalfe’s
Law13, this implies that once certain digital patterns of interaction become dominant
— such as the protocols that shape internet communication or define how computers
structure information — the social interpretation or regulation of these technologies

152
is already limited. Furthermore, informational production, by its structure, tends to
produce monopolies of both standards and firms, thus creating not only framed patterns
of interaction but also extreme power inequalities with the network structures.
Analysis has tended not to recognise the central role of domain knowledge and
its transformation on the micro-level within information technology. The informational
aspect of IT signals not only communication but also the embodying of social practices
and interactions in technology itself. These basic aspects, the micro-foundations and
patterns of informational processes, pose the crucial questions surrounding issues of
IT and development. E-commerce as a derivative of IT will impact upon economies
and societies even if the effects are only secondary rather than through primary
participation. Understanding the nature of this impact, and how societies and
governments can negotiate such interactions, is a central policy question. Even as (or
if) e-commerce becomes widely established, the domain-specific nature of
informational processes and products ensures that it will reflect a myriad of local
patterns, capacities and cultures that will be impossible to predict. How these local
patterns are tied into broader networks, however — and more importantly how general
informational development patterns may impact upon local communities — can be
understood and planned for through a careful consideration of the micro-foundations
of information societies.

153
 Notes

1. Hardware, like software, is structured around algorithmic codes that determine the
basic circuitry of an IC. In this way, hardware and software both follow the basic
structures of binary, algorithmic logic. The real choice is between whether such
patterns are “hardwired”, with linked tradeoffs between flexibility, reliability,
adaptability and cost. Software increasingly exceeds hardware in each of these areas
(Armour, 2000).
2. A fuller analysis of the contrast between hardware and software and the implications
for development can be found Eischen (2000a).
3. “Wrong Solution”, by Kumar Venkat, San Jose Mercury News, 8 April 2001.
4. See the Global Networks, Innovation and Development: The Informational Region
as a Development Strategy conference web site, http://www2.ucsc.edu/cgirs/
conferences/globalnet/index.html, for an overview of these issues drawn from
specific regional case studies.
5. See Eischen (2001) for a more detailed analysis of the development of the information
model and patterns.
6. This section is derived in part from fieldwork conducted in 1999/2000.
7. Andhra Pradesh’s average household income is $600 a year (four times the national
value); 63 per cent of the households have electrical connections and 21 per cent of
the inhabitants earn less than $49 per year (the Andhra Pradesh poverty line).
8. Interview: Joint Director Kumar, Software Technology Parks, Hyderabad.
9. See Bajpai and Radjou (1999) and Bajpai & Dokeniya (1999) for a comparative
analysis of Tamil Nadu’s IT strategy, and Parthasarathy (1999) on Bangalore.
10. See Lubeck & Eischen (2000) on Mexico, O’Riain (1999) on Ireland and Rasiah &
Best (2000) on Malaysia for comparison.
11. While not large compared with VC funds in the United States (which can be as large
as a billion dollars), 15 Crore is large by Indian standards. The federal government
VC fund designated for all of India has only 100 Crore in funding. To match the
same relative per capita commitment as the regional fund, the national fund would
have to be 12.5 times as large.

154
12. While outside of the analysis here, the social and economic patterns surrounding IT
industries such as software are clearly visible in other IT regions and industries,
including Los Angeles (film), San Francisco (biotechnology and e-commerce), Austin
(computers and ICs), New York (finance), Bombay (film) and Hong Kong (finance,
e-commerce, film).
13. Metcalfe’s Law states that the “value” or “power” of a network increases in proportion
to roughly the square (for large numbers) of the number of nodes on the network, or
more precisely N2-N.

155
 Bibliography

AMIN, A. (ed.) (1994), Post-Fordism: A Reader, Blackwell, Cambridge, MA.

ANGEL, D.P. (1994), Restructuring for Innovation: the Remaking of the U.S. Semiconductor
Industry, Guilford, New York, NY.
ARMOUR, P.G. (2000), “The Case for a New Business Model”, Communications of the ACM,
Vol. 43, No. 8.
BAJPAI, N. and A. DOKENIYA (1999), “Information Technology-Led Growth Policies: A Case
Study of Tamil Nadu”, Development Discussion Paper 729, Harvard Institute for
International Development.
BAJPAI N. and N. RADJOU (1999), “Raising the Global Competitiveness of Tamil Nadu’s
Information Technology Industry”, Development Discussion Paper 728, Harvard
Institute for International Development.
BENNER, C. (2000), “Navigating Flexibilities: Labor Markets and Intermediaries in Silicon
Valley”, doctoral dissertation, unpublished, Department of Urban Studies and
Regional Planning, University of California, Berkeley, CA.
BORRUS, M. (1993), The Regional Architecture of Global Electronics: Trajectories, Linkages
and Access to Technology, Berkeley Roundtable on the International Economy,
University of California, Berkeley, CA.
BROWN, J.S. and P. D UGUID (2000), Social Life of Information, Harvard Business School
Press, Boston, MA.
CAIRNCROSS, F. (1997), The Death of Distance: How the Communications Revolution will
Change Our Lives, Harvard Business School Press, Boston, MA.
CASTELLS, M. (1996), The Information Age: The Rise of the Network Society, Blackwell,
Cambridge, MA.
C ASTELLS , M. (1989), The Informational City: Information Technology, Economic
Restructuring, and the Urban-regional Process, Blackwell, Oxford, UK.
CASTELLS, M. and P. HALL (1994), Technopoles of the World: The Making of 21st Century
Industrial Complexes, Routledge, New York, NY.
D ERTOUZOS, M.L. (2001), The Unfinished Revolution: Human-Centered Computers and
What They Can Do for Us, HarperCollins, New York, NY.

156
DE V OL, R.C. (2000), Blueprint for a High-Tech Cluster: The Case of the Microsystems
Industry in the Southwest, Milken Institute, www.milken-inst.org.
DE VOL, R.C. (1999), America’s High-Tech Economy: Growth, Development, and Risks for
Metropolitan Areas, Milken Institute, www.milken-inst.org.
DORAY, B. (1988), From Taylorism to Fordism: a Rational Madness, Free Association Books,
London.
DOSI, G. (1984), Technical Change and Industrial Transformation, St. Martin’s Press, New
York, NY.
DRUCKER, P.F. (1993), Post-Capitalist Society, HarperBusiness, New York, NY.
EISCHEN, K. (2001), “Mapping the Micro-Foundations of the Informational Society: Linking
Software Processes, Products and Industries to Global Trends”, Center for Global,
International and Regional Studies, University of California, Santa Cruz, CA.
EISCHEN, K. (2000a), “Information Technology: History, Practice and Implications for
Development”, CGRIS Working Paper 200-4, Center for Global, International and
Regional Studies, University of California, Santa Cruz, CA.
EISCHEN, K. (2000b), “National Legacies, Software Technology Clusters and Institutional
Innovation: The Dichotomy of Regional Development in Andhra Pradesh, India”,
paper presented at the Association of Collegiate Schools of Planning Annual
Conference, 2-5 November 2000, Atlanta, GA.
GEERTZ, C. (1983), Local Knowledge: Further Essays in Interpretive Anthropology, Basic
Books, New York, NY.
GIORDANO, L. (1992), Beyond Taylorism: Computerization and the New Industrial Relations,
St. Martin’s Press, New York, NY.
GORDON, R. (1994), “State, Milieu, Network: Systems of Innovation in Silicon Valley”,
Working Paper 94-4, Center for the Study of Global Transformations, University of
California, Santa Cruz, CA.
GOVERNMENT OF ANDHRA PRADESH (2000), AP First: Information Technology Policy.
HELD, D., A. MCGREW, D. GOLDBLATT and J. PERRATON (1999), Global Transformations: Politics,
Economics and Culture, Stanford University Press, Stanford, CA.
HENDERSON, J.W. (1989), The Globalisation of High-Technology Production: Society, Space,
and Semiconductors in the Restructuring of the Modern World, Routledge, New
York, NY.
JURGENS, U., T. MALSCH and K. DOHSE (1993), Breaking from Taylorism: Changing Forms of
Work in the Automobile Industry, Cambridge University Press, London.
KRANZBERG, M. (1985), “The Information Age: Evolution or Revolution?”, in B.R. GUILE
(ed.), Information Technologies and Social Transformation, National Academy Press,
Washington, D.C.
LASH, S. and J. URRY (1994), Economies of Signs and Space, Sage, Thousand Oaks, CA.
LESSIG, L. (1999), Code: and Other Laws of Cyberspace, Basic Books, New York, NY.

157
LUBECK, P. and K. EISCHEN (2000), “Silicon Islands and Silicon ‘Valles’: Informational
Networks and Regional Development in an Era of Globalization”, in N. KLAHN, P.
CASTILLO, A. ALVAREZ and F. MANCHÓN (eds.), Las Nuevas Fronteras del Siglo XX1/New
Frontiers of the 21st Century, La Jornada Ediciones, México, D.F.
MALECKI, E. and P. OINAS (eds.) (1999), Making Connections: Technological Learning and
Regional Economic Change, Ashgate, Brookfield, VT.
MITCHELL, W.J. (1998), City of Bits: Space, Place and the Infobahn, MIT Press, Cambridge,
MA.
M YTELKA, L.K. (1999), Competition, Innovation and Competitiveness in Developing
Countries, Development Centre Studies, OECD, Paris.
NASSCOM (2001), Internet & E-commerce — The Indian Scenario, New Dehli, National
Association of Software & Service Companies, August, www.nasscom.org/it_industry/
int_ecomm_home.asp
NEGROPONTE, N. (1995), Being Digital, Alfred A. Knopf, New York, NY.
O’R IAIN, S. (1999), “The Flexible Developmental State: Globalization, Information
Technology and the ‘Celtic Tiger’”, conference on Global Networks, Innovation
and Regional Development: The Informational Region as Development Strategy,
11-13 November, University of California, Santa Cruz, CA.
PARTHASARATHY, B. (1999), “Institutional Embeddedness and Regional Industrialization:
The State and the Indian Computer Software Industry”, conference on Global
Networks, Innovation and Regional Development: The Informational Region as
Development Strategy, 11-13 November, University of California, Santa Cruz, CA.
RASIAH, R. and M. BEST (2000), “Industrial Transition in the Malaysian Electronics Industry”,
unpublished working paper, United Nations Industrial Development Organization,
United Nations Development Programme.
RHODES, R. (ed.) (1999), Visions of Technology: A Century of Vital Debate About Machines,
Systems and the Human World, Simon & Schuster, New York, NY.
R OSENBERG, N. (1982), Inside the Black Box: Technology and Economics, Cambridge
University Press, New York, NY.
San Jose Mercury News (various), www.sjmercury.com.
SAXENIAN , A.L. (1999), Silicon Valley’s New Immigrant Entrepreneurs, Public Policy
Institute of California, San Francisco, CA.
SAXENIAN, A.L. (1994), Regional Advantage: Culture and Competition in Silicon Valley
and Route 128, Harvard University Press, Cambridge, MA.
SHAPIRO, C. and H.R. VARIAN (1999), Information Rules: A Strategic Guide to the Network
Economy, Harvard Business School Press, Boston, MA.
SHELDRAKE, J. (1996), Management Theory: from Taylorism to Japanization, International
Thomson Business Press, Boston, MA.
US DEPARTMENT OF COMMERCE (2000), Digital Economy 2000, Economics and Statistics
Administration, Washington, D.C.

158
WARING, S.P. (1991), Taylorism Transformed: Scientific Management Theory since 1945,
University of North Carolina Press, Chapel Hill, NC.
WEBSTER, F. (1995), Theories of the Information Society, Routledge, New York, NY.
WOMACK, J.P., D.T. JONES and D. ROOS (1991), The Machine that Changed the World, Harper
Perennial, New York, NY.

159
 Chapter 8

Cellular Phones in Rural Bangladesh:

A Study of the Village Pay Phone of Grameen Bank
Salahuddin M. Aminuzzaman*

Introduction

Bangladesh suffers one of the lowest telephone penetration rates in the world,
even the lowest in the South Asian region. It has only three fixed and mobile phones
per 1 000 people, compared with five in Nepal, 15 in India, 20 in Sri Lanka and 25 in
Pakistan1. Most of the rural population is poor and cannot afford subscription to fixed
telecommunication lines. About 95 per cent of Bangladesh’s villages thus lack any
access to a phone. Moreover, the start-up cost of a fixed-line telephone in Bangladesh
is the highest in South Asia2 and involves a long waiting period of 5-10 years3.
In March 1997, Grameen Telecom (GT)4, a sister company of Grameen Bank
(GB), launched its cellular operations in rural Bangladesh. Twenty-eight GB borrowers
became pioneering operators of telephone services in their villages. By early 2001 the
number reached 1 500. GT plans to provide GSM cellular phone service to 100 million
rural inhabitants over the next four years in Bangladesh’s 68 000 villages. The
expansion programme will be carried out by GB micro-credit network. GT also aims
to organise the Village Pay Phone (VPP) service further as an income-generating
project for village operators.
This chapter describes the VPP system and presents the results of a unique study
to assess its efficacy in reducing the information poverty of villages that have obtained
access to mobile phones. More specifically, the study sought to analyse the ways in
which the VPP is operated, how the service is used and by whom and its impact on the
economic and social empowerment of individuals and communities5.
______________________________
* This study was supported by a research grant of NORAD and Telenor, Norway. In addition
to the author, the research team was composed of Prof. Harald Baldersheim, Department
of Political Science, University of Oslo, and Dr. Ishtiaq Jamil, Associate Professor,
Department of Administration and Organisation Theory, University of Bergen, Norway.

161
The Grameen Telecom System

To address the problem of poor telecommunications in rural areas, GT seeks to

introduce modern technology (cellular phones) that overcomes the lack of appropriate
public telephone services and the prohibitively high cost of traditional
telecommunications infrastructure. It uses GB loans to enable operators to underwrite
the initial investment of acquiring a handset for hiring out to local callers (the phone
and accessories cost an estimated $420). A GB client, most frequently a woman,
borrows about $300, then starts marketing the service to the village people. From the
ensuing profit, she earns a living besides paying back the loan. In selecting VPP
operators, GT follows certain criteria: at least two years as a GB member, a record of
on-time loan repayments, knowledge of English digits, a centrally located house or
place of business, one or two other sources of income and access to electricity (to
charge the phone). If qualified, the woman is lent the necessary capital at 22 per cent
interest with a three-year grace period. A VPP is normally installed at the operator’s
residence, although during the day it is placed in a tea stall or shop along a main
village road or close to the bazaar. The VPP operator charges the market retail rate
while GT charges her half that rate. The VPP owner initially uses a banner or poster
to draw the attention of potential users. In due course her house may become popular
and be known locally as Phone Bari (house with phone). During our field visits we
noticed that in a given village almost everybody irrespective of age could indicate the
Phone Bari or the VPP holder by name.

Objectives of the Study

The study on which this paper is based aimed at assessing the efficacy of the
VPP in reducing the information poverty of villages that have obtained access to
mobile phones. It analysed how the VPPs are operated, how the service is used and by
whom, and its impact in terms of the economic and social empowerment of individuals
and communities6.
The formulation of specific research questions was guided by the notion of
information poverty (Nafstad and Iversen, 1996). For present purposes, the term
designates a situation in which an inadequate telecommunications infrastructure — in
terms of quantity and price — limits the range of choices available to individuals by
making it too costly to seek out information about alternative courses of action. The
structure of hypotheses was guided by the theories of asymmetrical information and
cultural perspectives on collective action. Asymmetrical information is characteristic
of situations in which the quality of the object of the exchange is fully known to only
one of the parties. The inherent uncertainty that it generates may reduce the volume
of transactions that would otherwise take place. Sometimes the use of special
information is downright prohibited (insider trading). A number of social mechanisms
exist to rectify the problem and thus encourage transactions, e.g. the sale of goods and

162
services with guarantees attached, or the introduction of licensing or quality
programmes. A reduction of the asymmetry of information distribution should improve
the functioning of markets (Olsen, 1989). Asymmetrical information may also become
the basis of power relations. Such relations can be observed in an extreme form in
Bangladesh, where an illiterate majority depends on transactions with a literate minority.
Cellular phones may have the potential to even out the distribution of information and
thus reduce such asymmetries.
Recent elaboration of the theory of collective action has emphasised the cultural
foundations of collective action. Trust has been shown to be a prerequisite for collective
action, i.e. for overcoming the free-rider problem (Putnam, 1993; Fukuyama, 1996).
Interpersonal trust has cultural roots. Societies vary in the inclination of people to
trust each other spontaneously. In low-trust societies collective action is difficult to
organise beyond the extended family or the village. Such societies tend to be fragmented
socially and highly state-centred. State authority provides a substitute for the social
cement of the mutual trust of civil society. The introduction of cellular phones in a
low-trust context should facilitate transactions that might otherwise not take place or
take place at higher costs. A cellular phone may enable the parties to check out on the
spot information given by the other and thus allay fears of being cheated.

Methodology and Analytical Model

The study was based on a sample survey of 350 VPP owners/operators and
users. Samples were drawn on the basis of geographical locations and time since the
introduction of VPP in the locality. In addition, a number of distant beneficiaries
(persons who had received calls from VPPs) were interviewed. To supplement the
survey some case studies were also carried out.
The basic research issue in the study concerns the identification of the processes
that influence the success of VPP operations. What makes for the success or failure of
a technology-driven development initiative such as the VPP? The approach adopted
was a demand-and-supply model supplemented with contextual analysis. One would
expect the intervention to succeed if it meets a pent-up demand, if the supply is
technologically adequate and if the costs fall within the customer’s willingness or
capacity to pay. Insofar as both demand and supply may be influenced by contextual
factors, however, such factors can also impinge upon the success of the VPP. They
may be related to the economic, social and cultural environment of the communities
into which the VPPs are introduced. Some economic activities may generate more
demand for phones than others: businessmen, for example, may ceteris paribus need
to make more calls than peasants. Communities with many migrant workers living
abroad may also grasp the opportunity for better communications more eagerly than
those with few such individuals. Some VPP holders may be more active in promoting
the service than others so that the supply side is different. The technology may work
better in some places than others for geographical reasons. Cultural factors may also

163
come into the equation. For example, more conservative communities may take longer
to get used to or accept new gadgets; they may also object to women as operators.
Both situations may hamper demand. Changing the supply side by letting men take over
the day-to-day operations may stimulate demand in such settings. The basic model is
sketched in Figure 8.1. Some key details of the model are listed below the figure.

Figure 8.1. An Analytical Model of the VPP Intervention

Supply Demand

3 1 2
Operations of VPP Success/Failure User need of VPP
Empowerment of users of VPP

4
Contextual factors

— Success and failure are measured in terms of users’ empowerment as indicated

in interviews.
— User need is an indirect measure built by classifying users into professional
categories (e.g. housewives, businessmen, peasants, etc.), as well as age and
education groups, that show different communication needs.
— VPP operations are not systematically analysed in relation to user empowerment
because observations during the study revealed them to be fairly standardised.
Some indications of technological and other problems in VPP operations are
reported, however, as are holders’ perceptions of the impacts of the VPP.
— The contextual categories are used mainly to distinguish between the Chittagong
and Dhaka zones. Chittagong is well known for its distinct regional culture. It is
both economically entrepreneurial and socially conservative. Thanks to its port
and a strong merchant tradition, as well as the unusually high proportion of
migrant workers, it is open to the world. At the same time, long-time traditions
are held in high esteem, as reflected for example in the preference accorded to
marriages within the region. Chittagong’s entrepreneurial character may be
expected to generate more communication needs and therefore higher intensity
in the use of the VPP. Its social conservatism, on the other hand, may constitute
an obstacle to accepting the VPP and its special features of operation, chiefly
female holders/operators.

164
Main Findings

Profile of the VPP Stakeholders

Generally, the users are younger than the owners. Around 75 per cent of owners
are between 35 and 55 years old, with the largest group in the 35-45 age bracket.
Among the users, the largest age group is 24-35 years old, with 19 per cent less than
25. Few users are above 55, so the use of the telephone is definitely something for the
younger generation. Most of the owners (91.8 per cent) are female while the majority
of the users (88 per cent) are male. All but one of the owners are married. Reflecting
the high marriage rate in rural Bangladesh, it is very rare for mature women to remain
unmarried. Fewer users (67.7 per cent) are married because they are also younger
Literacy among both owners and users is higher than average for rural Bangladesh.
Only 4 per cent of users and 12 per cent of owners are illiterate, as against an 80 per
cent general illiteracy rate for adults. More than 80 per cent of users have education
beyond the primary level, while among the owners 47 per cent have such schooling.
Both groups represent something of an educated elite in rural Bangladesh. The
difference in educational levels between users and owners reflects the contrast in
school attendance among boys and girls, the owners being mostly female and the
users mostly male.
Most of the owners (77.6 per cent) are primarily housewives engaged in household
activities. Only a few have other activities such as entrepreneurship (12 per cent),
trading (4 per cent), services (4.7 per cent) and agricultural work (2.4 per cent). On
the other hand, as many as 41.1 per cent of the VPP users are traders and 22.8 per cent
are small entrepreneurs. Some users also do household activities (11.4 per cent) and
some are service providers (13.3 per cent). Only a tiny proportion of users works in
agriculture (3.2 per cent).

Call Patterns

The average number of calls made in a month is 16, with a range from one to
well over 50. The most usual call frequency is in the range of 11 to 25 calls every
month, signifying a good number of users of the VPP service that make calls almost
every day or every second day. This pattern clearly demonstrates that the VPP has
identified a need and effective demand arising out of that need. An analysis of the call
frequency of various occupational groups indicates as expected a conspicuous contrast
between them (Figure 8.2). Traders, service providers, entrepreneurs and students
most frequently use the VPP, while those in agricultural occupations and housewives
do so more intermittently. Calls therefore are probably driven mostly by professional
and economic motives. Social motives, however, top the list of reasons for calling.
Calls to family members and relatives dominate among them. Business calls come
second, while official (government-related) and political errands are quite rare. The
last may reflect the limited role that the government plays in the life of the villages.

165
 Figure 8.2. Distribution of Average Monthly Call Frequency by Occupation
20

18
Average monthly call frequency

16

14

12

10

4
Household Trading Entrepreneur Service Agriculture Student
activities & other
Occupation of the user

Business-Related Impacts

The study sought to identify the impacts of the VPP on individual and community
life. “Impact” refers to the difference that access to the VPP has meant to the individuals
in the study areas. Assessments of impacts are based on the self-reported advantages
of VPP access that the interviewees indicated. Users reported a number of business-
related advantages (Table 8.1). Access to the VPP has meant above all the creation of
more opportunities and choices, but it has also provided help in managing uncertainty.
Moreover, existing business relations have been strengthened.

Table 8.1. Business-Related Advantages of VPP

Percentage
VPP has helped in
of Users

Providing more choices for decision making 78.50

Taking uncertain decisions 68.40
Strengthening relationships with business partners 59.50
Motivating the user to take new initiatives 53.80
Creating new economic and income-generating opportunities 40.50
Note: The percentages are based on cases, not on responses. Because of multiple responses, the total
exceeded 100 per cent.

166
 Box 8.1. The Voice of a User
I call mostly for business purposes. I sell eggs to a middleman. Before having VPP, I
had to accept the middleman’s price offer for the eggs because I had no other way to
know the market price for eggs in the bigger markets. VPP gave me the opportunity to
verify the market price of eggs. Now, before I sell eggs to middlemen, I do verify the
market price in the nearby markets and I agree to sell only when I get a good price.
Now I feel much more confident as I have gained bargaining power with the middlemen
who mostly deprive us of our profit.

The advantages that the users feel the VPP has given them in business transactions
relate above all to the reduced access time to get information (Table 8.2). Many also
mention an enhanced feeling of security by being able to convey messages rapidly.
Reduced communication expenses are also important.

Table 8.2. Advantages of VPP to Users as to Collection of Information

Percentage
Advantages
of Users
Reduced access time to information 44.1
VPP has become an information centre, can easily be shared 17.9

Reduced feeling of physical distance to the information centre 29.6

Reduced communication expenses 27.9
Negates physical presence 26.4
More security through transmitting the message instantly 39.7

Note: Percentages are based on cases, not on responses. Because of multiple responses in each case, the total
exceeded 100 per cent.

Social and Family Relationships

In rural areas people remain isolated from their family members and friends
who do not live in nearby places. Also, because Bangladesh is a manpower exporting
country, as many as 3 million Bangladeshi, mostly semi-skilled and unskilled labourers,
live abroad, many of them in the Middle East and Far East. A good number of non-
resident Bangladeshi also work both legally and illegally in various European and
North American countries. These expatriate Bangladeshi and their relatives stay in
touch mostly through letters. As most people in rural areas are not literate, they depend
on others to write and read letters for them. Hence, they cannot maintain privacy and
are prevented from instant sharing of joys and sorrows. VPP has the potential to
provide emotional relief for people by allowing them to contact their loved ones in
situations of privacy. VPP can also potentially help in situations requiring medical

167
advice and assistance. Furthermore, rural women are more isolated than men in
Bangladesh because of the traditional purdah system (seclusion of women from men).
After marriage, many women cannot visit their families because of the prevailing,
male-dominated cultural practices. They are expected to concentrate on their household
duties, and they themselves cannot take the decision to visit their parents because their
husbands and in-laws exercise authority over them in such matters (Firdhouse and
Zamen, 1994). In such situations, the VPP represents an opportunity for a woman to
maintain contacts with parents, brothers and sisters.
VPP has been extensively used for making social calls. VPP owners have reported
that social calls7 dominate both incoming calls (64.22 per cent on average) and outgoing
calls (61.44 per cent)). In many cases these social calls also have business purposes.
When a migrant labourer calls to his family, in most cases the call includes a message
related to monetary transactions, i.e. sending or receipt of money, purchase of livestock
or land, construction of a house, etc. In response to how VPP affected family
relationships, user opinions were distributed along three dimensions. The calls:
— helped in easier sharing of private emotions (36.9 per cent) by providing more
privacy than other modes of communication;
— strengthened family relationships by reviving broken or forgotten relationships
(34.4 per cent); and
— reduced suspicions, which again led to better relationships (30.3 per cent).

Changes in Conceptions of Time and Space

Most users (96.2 per cent) reported having saved time through VPP. Savings
seem to have been substantial for most purposes for which the telephone was used.
Most users (96.2 per cent) also stated that VPP has enabled them to save on
transportation costs. For the majority of the users (71.7 per cent) transportation cost
savings have been up to Tk. 500 a month.
Nearly half of the users (48.7 per cent) considered that VPP helped them to
prepare better for the natural disasters that so often can happen in Bangladesh, but
only 39.2 per cent have used VPP for seeking medical assistance (as against 71.8 per
cent of the owners). The low rate of medical-oriented use could reflect the nature of
health service-seeking behaviour of the rural population. People in the rural areas
generally visit traditional healers (spiritual healers, homeopathic doctors, village
quacks, etc.) rather than health service delivery points. Moreover, those points cannot,
as a rule, be reached via the phone.
As VPP has affected users’ and owners’ time and costs, it has also had an impact
on their transportation needs. Most of the users (91.1 per cent) and owners (95.3 per
cent) confirmed that VPP has affected travel patterns. For the users, the VPP’s impact
on transportation costs is significantly correlated with VPP impact on travel (r = 0.60,
sig. 0.01 level). Fewer trips mean less travel cost. Among the users, 38.2 per cent

168
stated that their frequency of travelling for business purposes was reduced and 22.9 per
cent said that they were not required to travel at all after access to VPP. Overall,
15.6 per cent of the respondents mentioned that their frequency of travel for official
purposes had been reduced.

Economic Empowerment

Empowerment is the reduction of dependency relations. The figures emphasise

the role of communication and information in producing alternatives and choice and
thus breaking cycles of dependency. The VPP is a channel for communication and
dissemination of information about alternatives. A by-product is the opportunity for
maintaining or widening social networks, which again may produce more information.
In rural Bangladesh, people have very little scope for choice in work or social relations
but remain confined to the village and its limited income-earning opportunities. The
scope for women is even narrower than that for men. Owners as well as users thus
experienced a variety of changes after their access to the VPP — changes more
pronounced for owners than for users, which may be natural because the VPP is a
direct income opportunity for the owners. Being mostly women, owners may also
have started from more restricted life situations than the users did. Both groups
experienced increased income, widening networks and more business. More detailed
information on this is presented below.
Economic empowerment refers not only to increases in income but also to having
control over resources and resource management, decision-making power and
involvement in and control over economic transactions. Obviously, VPP as a business
venture provides an opportunity for financial gain for the owners. Most of the owners
(94.10 per cent) found VPP profitable as an investment. VPP has brought significant
changes in the range of average monthly incomes of the owners (Table 8.3). In most
cases, owners’ incomes increased with the length of time of VPP ownership. Those
operating the VPP for 1-6 months had an increase in average income of Tk. 2 204.55,
those in operation for 6-12 months had a Tk. 3 453.49 gain and owners in the VPP
business for 12-18 months enjoyed an average income Tk. 3 125 higher. There was
also a substantial increase for the 24-30 months group. Apparently the VPP has been
a success for the owners as an income opportunity.

Table 8.3. Distribution of Owners’ Average Monthly Incomes before and after Owning VPP

Average Monthly Income (Tk.) Percentage of Owners

Before VPP After VPP
1 500 and below 19.7 7.4
1 500–3 000 33.3 12.3
3 000–6 000 23.5 33.3
6 000–9 000 8.6 22.2
9 000 and above 14.8 24.6

169
 Social Empowerment

VPP has had a remarkable effect in widening the networks of the beneficiaries.
Most of the users (95.6 per cent) spoke in favour of VPP as helping to widen their
contact/clientele groups. Almost two-thirds (63.2 per cent) stated that VPP helped to
a great extent. The VPP’s impact on the expansion of users’ social circles is significantly
correlated with their increase in incomes (r = 0.31, sig. 0.05 level). Obviously, gains
in social capital relate to financial gain. A similar pattern appeared among the owners.

Box 8.2. Another User’s Voice

Use of VPP has led to a significant widening of my business network. Now I can
communicate with many people through someone else’s reference and can easily
establish business relationships. Before VPP, it was not always possible to establish
contact with people because in many cases it was not cost effective for me.

In rural Bangladesh, owning a phone is a symbol of social prestige and power,

not because it relates to income-earning power but because owning a VPP in itself
bears the mark of social prestige and honour. As mentioned earlier, the house of the
VPP owner is commonly named as Phone Bari (house with phone). In some cases the
VPP owner is also lightly called Phone Bibi (lady with phone). Most of the owners
(92.9 per cent) held the opinion that owning a VPP did bring them social prestige. A
significant proportion (39.2 per cent) said that because of VPP they have become
“reference personalities” in their villages.

Change in Bureaucracy-Citizen Relationships

People may have to contact public offices and officials for many different
purposes, including dispute reporting, information gathering, official reporting,
complaining, tadbir8, political influencing, etc. In Bangladesh, getting a job done or
collecting any information can be time-consuming, particularly in public offices. People
may have to go to public offices or contact public officials repeatedly even for small
errands. It is more difficult for poor people without good connections. The introduction
of VPP has made it easier to reach the bureaucracy. Through the VPP users can
communicate more easily with the Thana (the local police station), government officials
and political leaders, especially members of parliament. They do call them for varied
reasons. Comparatively more calls are made to government officials and political
leaders than to the Thana. Dispute reporting dominates among the calls made to the
Thana (57.1 per cent). Information collection and official reporting/interaction
predominated among those made to GOB officials (28.9 per cent and 26.7 per cent
respectively). Tadbir and information collection were the main objectives of calls
made to political leaders (34 per cent and 36.2 per cent respectively).

170
 Women’s Empowerment

VPPs have created an income-generation opportunity for rural women and

provided scope for interacting with a wider cross-section of people. Before they had
VPP, women’s access to phones was difficult. To assess if VPP could, aside from
financial gain, also facilitate the economic empowerment of women, the study probed
not only VPP owners’ feelings about their empowerment, but also the users’ perceptions
of it. When asked about the VPP’s role in empowering women, most owners (81 per
cent) stated that VPP had added to their social prestige. That they are known as VPP
owners has indeed changed their social image. To them, ownership itself is a matter of
social dignity although in most cases they are not responsible for the day-to-day
operation of the VPP.
From the point of view of women’s economic empowerment, it is also vital to
take into account who is responsible for the actual operations of the VPP. Because of
the very nature of Grameen Bank policy, most of the VPP owners are female (91.8 per
cent). Yet the female owners operated only 4 per cent of the VPPs themselves. Male
family members of the female owners operated 96 per cent of the VPPs. Such exclusion
from the VPP practicalities may lead to women’s exclusion from decision making
involved in VPP operations and from the use of income from VPP businesses. When
owners are not involved in VPP transactions, they are probably also less exposed to
social interactions. Moreover, as mentioned above, only 22 per cent of VPP users are
female. It seems, therefore, that in rural Bangladesh, in both the use and operation of
the VPP, the traditional seclusion of women from the wider society has been maintained.
Apparently, tradition and culture influence both the demand and supply side of the
VPP. Nevertheless, for a significant minority of women, the VPP has entailed some
change. Two-thirds of the users said that VPP had an impact on women’s
empowerment, but most of them also felt that introduction of VPP had not caused
any change in men-women relationships. Only 25.9 per cent thought they had noticed
any change in this respect.

Impact on the Distant Beneficiary and User

It has been assumed that, besides affecting owners and users, the introduction of
the VPP also affects people who are the regular contact persons of the VPP users. The
study termed them ”distant beneficiaries” or “distant users” of VPP phones. Distant
beneficiaries include traders, service owners, students, local political leaders and
housewives. All of them said that they had indeed benefited from the introduction of
the VPP. The most commonly stated benefits were reduced communication time
(100 per cent) and saved communication cost (88 per cent). Significant percentages
also felt that the VPP had extended their communication networks (36 per cent) and
widened their business networks (66 per cent). Before the introduction of VPP,
communicating with the present users was a cumbersome process. The distant
beneficiaries had to go to them physically or send representatives. Sending letters
took a long time. Most of the distant beneficiaries (80 per cent) often sent

171
representatives. With VPP, 28 per cent of the distant beneficiaries received between
one and five calls from the users per week and 44 per cent made one to five calls to
the users per week.

What are the Most Important Impacts of the VPP?

The VPP has entailed a series of changes in social and economic relations. So
far in this study a quite rich and varied picture has emerged. A series of impacts has
been identified, most of them in the expected direction. It is more difficult, however,
to assess the relative magnitude and significance of the various changes and impacts
from the detailed analysis. This section tries to arrive at a more comprehensive
understanding of the impacts of the VPP and the forces that help shape them.
To obtain a summary picture, all the variables seeking to identify such impacts
were re-coded into dichotomous variables. They were then subjected to factor analysis.
The factor solution yielded seven different factors, which were used as guidelines for
the construction of indices of impact. These indices can also be interpreted as dimensions
of empowerment associated with the VPP. The indices were standardised to a scale of
0-100. Figure 8.3 shows the results of the analysis. Note that it is based on data about
users only.

Figure 8.3. Impact of Cellular Phones as Assessed by the Users

Impact areas
100
90

80

70
60
Intensity

50

40
30

20
10
0
Economic Political/ Social Transport/ Women’s Family Uncertainty
Administrative Mobility Development Index Management

The strongest impacts of the VPP appeared for transport and mobility and family
relations. The VPP enhances the geographic reach of the users while at the same time
reducing the need to travel to obtain information. These, of course, are quite obvious
effects once the VPP becomes operational in an area, but the ability to make the VPP

172
operational could not be taken for granted and data therefore were collected on the
seemingly “obvious” aspects of the telephone’s benefits. Significant effects also
emerged with regard to the social life and status of users, women’s empowerment
and uncertainty management in general. Surprisingly, the VPP has had less impact
on the economic activity of the users. Recall, however, that the direct economic
benefits were more pronounced for owners than for users, for whom the economic
benefits were primarily indirect.
Overall, then, the VPP allows a more efficient use of time and enhancement of
social and family relations by making it possible to stay in touch more frequently and
more directly. This is no small advantage in a family-oriented society such as
Bangladesh. Easier access to information through the VPP also reduces uncertainties
and anxieties in daily life and helps quicker and better decisions in the various
occupations of the users. This must be regarded as a step towards a more efficiently
functioning society. It has also meant an increase in the volume of social and economic
transactions, as predicted by the theory of information asymmetry as well as that of
social capital.

Explaining Variations

The preceding sections have noted variations in the uses and impacts of the VPP.
They have suggested a number of supply-side, demand-side and contextual factors
that may account for such variations. Their impacts are analysed in Table 8.4. Reading
the table vertically, the columns show how the factors (independent variables) affect
the level of users’ empowerment in particular impact areas, for example family relations.
The higher the beta coefficients, the stronger the impact. Reading horizontally, a row
shows the impact that a given factor has across different impact areas.
For the overall impact (general index) summarised for all areas, the driving
forces are the ages and occupations of users. Younger users and those occupied as
traders have particularly felt the advantages of the VPP. Both of these effects seem
intuitively reasonable. Younger people are prone to make use of new gadgets more
quickly than their elders. Villagers working as traders will likely have a greater need
to get in touch with customers and suppliers easily and may hence welcome phones
even more than those in other walks of life.
The factors identified here do not so well explain the transport and mobility
impacts, perhaps because the impacts are of a rather general nature and therefore felt
equally by most users. Family-oriented impacts are felt less by the entrepreneurs than
by those in other occupations, and they are more manifest in the Chittagong zone than
in the Dhaka zone. This may reflect the larger number of migrant workers from
Chittagong, using the phone to be in touch with their families. It may also be a
manifestation of the strong regional identity of the region and therefore the strong
pull of regional roots.

173
 Table 8.4. Factors Affecting Impacts of VPP on Users: Regression Analysis.
Impact Areas (Dependent Variables)
Factors
(Independent Variables) General Impact Transport and Family Social Women Uncertainty Economic Political &
Index Mobility Management Administrative
Age of user -.258**

Education of user -.339*** E***

No. of family members of

user
Monthly family income of
user (in Taka)
Distance from the nearest
Thana town (in km)
Occupation:
Household -.481***
Occupation:
Trading .284** -.299** .342***
174

Occupation:
Entrepreneur -.250** -.310**
Occupation:
Service -.344**
Occupation:
Agriculture
Occupation:
Student
Zones: Chittagong=1 -.272** -.233**
Dhaka=2
Call frequency (monthly) .284** .277**
Period used VPP (months) .329***
Adjusted R2 .214 .052 .135 .012 .275 .070 .186 .120

Note: Levels of Significance: * = .05 level, ** = .01 level, *** = .001 level.
 Women’s empowerment is driven by education, occupation, zone and practice
as a VPP user. Surprisingly, the VPP is seen as bringing women’s empowerment more
by those with low education than those with higher education. Those working outside
household activities see women as more affected than those who are primarily occupied
in the home. Again, Chittagong dwellers are more prone than the Dhaka dwellers to
see women’s empowerment coming from the VPP. Such a perception has a lot to do
with how long the user has had access to the VPP. The impact of (low) education may
indicate that the VPP will likely make more of a difference to people with low education
than to those with higher educational achievements, who also have access to information
through other channels. The negative perceptions of householders (mostly housewives)
on the VPP’s impact on female empowerment may suggest that individuals confined
to the house have more difficulty in seeing the wider benefits of the phones. Yet they
may perhaps also tend to generalise from their own confined existence to women’s
empowerment in general. Furthermore, women’s empowerment tends to be emphasised
more the longer the user has had access to a VPP, which can be given a quite
straightforward interpretation.
Enhancement of economic opportunities is related to call frequency and
occupation. Traders, primarily, feel they have become more economically empowered
by the VPP. These are demand-driven relationships. Empowerment in relation to
political and administrative bodies relates to education levels in the expected way.
Those who use the phone more to get in touch with Thana, administration and political
leaders have more education than those who don’t.
Reading the table horizontally, it is apparent that no single factor or group of
factors explains variations in impact equally across the board, in all impact areas.
Occupation is an important factor in the call patterns — some occupations (especially
traders) have more need of the phone than others. In some areas age and education
turned out to be important but not ubiquitously so. Such presumably demand-driving
factors as family income and distance from nearest centre play very little role. The
same is true of family size. All in all, the explanatory power of these regressions may
not be too impressive (as indicated by the row of adjusted R2s), but it did help to
explain variations in some cases. It worked best for female empowerment and
economic enhancement.

Conclusions

Any large-scale mobile phone programme like Grameen Telecom’s Village Phones
will bring changes to the village economy and village life in general. The VPP has
expanded considerably the transaction boundaries of the local village market, reducing
the isolation and fragmentation of the village economy. GB’s style of managing
communications can help significantly to expand access to this vital information input
for all segments of the population, reduce inequality and thus enhance the broad-

175
based, pro-poor orientation of rural development strategies. There are significant
non-economic benefits, such as rapid and effective communications between family
members, stronger kinship bonding, etc.
Access to mobile phones has had a noticeable effect on development and
entrepreneurial activity in rural Bangladesh. Effective demand exists and appears to
be growing. The mobile phone has also brought some attitudinal change in social
acceptance of women as lead agents in creating conditions appropriate for the spread
of information technology. The experience of Grameen Telecom in Bangladesh indicates
that mobile phones can deliver telecommunications to impoverished rural areas
efficiently and profitably. It has also demonstrated that in a technology-shy society
like Bangladesh, a high technology can indeed find use, provided it is channelled
through a well designed institutional framework that does not disturb social norms
and values.

176
 Notes

1. Grameen Phone (1999), Annual Report.

2. Grameen Phone (1999), Annual Report, p. 44.
3. Grameen Phone (1999), Annual Report, p. 44.
4. Grameen Telecom is a not-for-profit company registered under the laws of Bangladesh
for attaining certain social objectives. In its efforts to extend the village phone
service in rural areas, it has the advantage of being assisted by its peer, Grameen
Bank, an internationally recognised bank for the poor with an extensive rural network
and understanding of the economic needs of the rural population. Grameen Telecom
administers the village phone services to villagers and trains the operators as well as
handling all service-related issues.
5. At a general level, Coates and Jarrat (1990) have suggested a series of typical social
impacts of mobile phones: acceleration of transactions, thereby providing
competitive advantages; expansion of scope of short-term transactions, e.g. the
convention of on-the-spot meetings; production of more choice (a buyer may for
example compare prices in one store with those of another with a call from the first
location); provision of opportunities for immediate feedback, e.g. phone voting;
stimulation of certain kinds of actions (both desirable and undesirable); increase in
public and individual safety; alteration of concepts of time, space and distance;
raising expectations.
6. See note 5.
7. Social calls refer to calls made or received by users for exchanging greetings between
and among friends, relatives and family members, as well as for knowing their
whereabouts.
8. That a client must employ extensive persuasion by using social and political links
to get things done in a public office is an element typical of a bureaucratic culture.

177
 Bibliography

COATES, J.F. and J. JARRAT (1990), “Future Use of Cellular Technology — Some Social
Implications”, Telecommunications Policy, Vol. 14, No. 1.
FIRDOUSE A. and N. ZAMAN (eds) (1994), Infinite Variety, UPL, Dhaka.
FUKUYAMA, F. (1996). Trust. The Social Virtues and the Creation of Prosperity, The Free
Press, New York, NY.
NAFSTAD, H. and A.W. IVERSEN (1996), “The Role of Information Technology in African
Development”, Telektronikk, Vol. 1, No. 1.
OLSEN, T.E. (1989), “Effektiv ressursallokering under asymmetrisk informasjon” (Effective
Resource Allocation Under Information Asymmetry), LOS-senteret, Bergen.
PUTNAM, R. (1993), Making Democracy Work. Civic Traditions in Modern Italy, Princeton
University Press, Princeton, NJ.

178
 Chapter 9

Local Content Creation and E-Commerce:

A South African Perspective
Carey-Ann Jackson and Johan Eksteen

Introduction

During the recent corrections in the technology sector across the world’s stock
markets, it became clear that the “click-and-mortar” model is a more robust business
model in the e-commerce arena. This model sees e-commerce as an extension and
enabler of existing, non-virtual businesses. Lessons learned from these events can
apply directly to the area of local content creation and e-commerce, and it is this
learning which informs the subject of this paper.
In discussing this subject, the paper recognises the need to address local content
creation and e-commerce as a holistic value chain located in a global environment.
The paper presents a view that such a value chain spans both virtual and physical
dimensions and stretches across a wider range of activities than just consumption
activities. Further, local content and e-commerce value chains have a number of
different channels, infrastructures, drivers and enablers, which run along the length
of the chain and differ considerably between one point and another.
In contrast with this approach, discussions of local content creation and e-
commerce commonly refer only to considerations of consumption and the market,
typically located at the latter parts of value chains. In promoting another option for
developing-country participation in global e-commerce, there is a need to understand
and support the whole value chain and in so doing account for differences of
environment, channels, infrastructures, drivers, enablers, potential risks and threats.
The paper refers to pilot research and development projects in South Africa and uses
these projects to illustrate what environmental conditions and associated enablers come
into play in the early and middle parts of the value chain.

179
The Argument

The approach to local content creation and e-commerce outlined in the paper
emphasises the need to think about these issues as a holistic value chain. The argument
begins with reference to the realities faced by all South African initiatives addressing
local content creation and e-commerce: the environment is dominated by the presence
of poverty and inequities in access to information and communications technologies
(ICTs), and these realities provide the background for the paper’s arguments about
local content creation and e-commerce.
Poverty alleviation is one of the highest developmental priorities in South Africa.
Almost half of all South Africans live in impoverished households (Statistics South
Africa, 1997). Most poor South Africans live in rural areas (Table 9.1) and a significant
proportion live in households headed by women. Widespread poverty impacts directly
upon participation by South Africans in global markets and on the country’s presence
on the World Wide Web.

Table 9.1. Poverty in South Africa

Province Households living in poverty Individuals living in poverty Non-urban population
(%) (%) (%)
Eastern Cape 40.4 50.0 63.4
Free State 56.8 64.0 31.4
Gauteng 29.7 41.0 3.0
KwaZulu-Natal 36.1 47.1 56.9
Mpumalanga 33.8 45.1 60.9
North West 15.4 21.1 65.1
Northern Cape 38.2 48.0 29.9
Northern Province 61.9 69.3 89.0
Western Cape 14.1 17.9 11.1
South Africa 35.2 45.7 46.3

Source: Adapted from Statistics South Africa 1997 data.

It would be remiss in a paper concerned with e-commerce not to describe access

to ICTs in South Africa. Common parlance recognises South Africa as a country with
a digital divide, yet some domestic commentators have criticised the use of the phrase
“digital divide” because it masks the social, economic, educational and commercial
inequities that predate the existence of anything digital. The provision of technological
infrastructure alone cannot eradicate these inequities and several strategies are needed
to do so, some of which are:
— strengthening and diversifying the human resource and competence base;
— ensuring participation by small, medium and micro enterprises (SMMEs) in the
economy;
— supporting job creation and black economic empowerment;

180
— providing impoverished households with access to finance;
— addressing the affordability of a range of basic services; and
— ensuring women’s empowerment.
Table 9.2 shows current access to telephones (including cellular or mobile
telephones). It exemplifies the disparities between two groups of South Africans (http:/
/www3.sn.apc.org/africa/). While “Africa’s information infrastructure is by far the
least developed in the world” (UNECA, cited in Hall, 2001), South Africa is
comparatively a continental “giant”, responsible for almost 95 per cent of all African
internet hosting. South Africa has over 100 000 hosting facilities, sophisticated IT
infrastructure in the finance and retail sectors, an advanced cellular communications
infrastructure and a growing community of internet subscribers — a profile similar to
those in some European countries (Hall, 2001) (Table 9.3). Several private South
African organisations have established web presence. Financial institutions, large
clothing and food chain stalls, general service providers and Internet Service Providers
(ISPs) provide on-line users with services such as electronic banking, booking and
reservation facilities, auctions, secure purchasing of anything from fast-moving consumer
goods (FMCGs) to motor vehicles, and job applications and recruitment, for example.

Table 9.2. South African Access to Telephones

(percentages)

All African White

Universal service 34 20 89
Universal access
(15 minute walk) 71 60 99
Universal access
(30 minute walk) 83 78 100
Universal access
(60 minute walk) 90 87 100
Source: Adapted from AISI data for South African access to telephones.

Table 9.3. Internet Access, Use and Service Provision in South Africa
Use, Access and Service Provision Frequency

Dialup subscribers 650 000

Leased lines 5 000
Estimated total users 1 800 000
Major dialup ISP Mweb
World Online
ABSA
Leased line & national network providers Internet Solutions
UUNet Africa
SAIX

Source: Network Wizards’ Internet Domain Survey.

181
 South Africa is expected to play a role in development and expansion of internet
connectivity across the Southern African Development Community (Table 9.4). Yet
“within South Africa, there are huge contrasts between the urban, largely white and
increasingly commercial users of information and communications technology, and
rural, overwhelmingly African, communities who have only partial access to basic
telecommunications” (Hall, 2001).

Table 9.4. Host Statistics in Egypt and Sub-Saharan Africa

Country Number of Hosts

South Africa 122 025
Egypt 2 013
Namibia 640
Zimbabwe 599
Botswana 550
Mozambique 69
Swaziland 330
Angola 4
Source: Network Wizards’ Internet Domain Survey

A word of caution is in order about describing access among individuals and

households in poor and rural communities in static terms. Most data sets lack reference
to demographic data on migration and mobility. Both are features of daily life,
particularly among impoverished individuals, and in South Africa large portions of
the population travel significant distances between their family households and working
(or work-seeking) places. This is a legacy of apartheid urban and rural planning as
well as a reflection of the social and economic changes underway in the society. Given
the frequency of significant movement of large groups of people in the society, describing
access (or the lack thereof) in statistical terms needs a clearer understanding of migration
and mobility data because they may, inadvertently, mask the real inequities of access
to infrastructure.
South Africa has implemented its ICT policies and regulations as attempts to
address the poverty and inequity that exist within the society. Through the provision
of an exclusivity period to the national telecommunications provider, the South African
government defined social responsibility rollout targets associated with extending
universal access. Similar social responsibility targets were included in the licences
granted to telecommunications players such as the public broadcasters and GSM service
providers.
A topical ICT debate underway in South Africa during the writing of this paper
concerns the formulation of the White Paper on e-commerce1. It is not uncommon to
hear argument that e-commerce in South Africa should be linked to poverty reduction
and wealth redistribution. Some community activists, leaders of organised labour,

182
policy makers and intellectuals want to ensure that e-commerce legislation is developed
so that benefits accrue for all. During the White Paper development process, Stavrou
et al. (2001) have promoted such an approach, inspired by the work of Amartya Sen,
that connects a livelihoods concern with e-commerce. Stavrou et al. (2001) highlight
the need to emphasise the assets and abilities of the poor and recognise that poor
households face a diversity of risks, such as natural and environmental hazards, price
fluctuations, policy changes, social relationship changes, conflicts and “unbanked”
money or cash management strategies. Risk aversion is widely attributed as characteristic
of poor households and the ways in which they manage their assets in relation to
perceived and actual risk. Those risk management strategies can, depending largely
upon the level of poverty of the household, result in a vicious cycle of poverty (Stavrou
et al., 2001). Their approach argues that access to information can support improved
risk management strategies and impact upon the household’s livelihood. Lack of
information or incorrect information can add substantially to poor households’ costs of
living compared with households that do have better informational access. Other key
points about the livelihoods approach, taken from Stavrou et al. (2001) are as follows:
— it recognises the possibility of exchange failure;
— it allows for separation of proximate resources for livelihood generation from
intermediary resources;
— it recognises that unwired households are disadvantaged economically further
by lack of connectivity;
— it emphasises diversification of income sources as crucial for sustainable
livelihoods; and
— it acknowledges that engagement with diversified activities permits households
to try different tactics for generating adequate and sustainable livelihoods.
This paper extends the livelihoods approach to e-commerce by arguing for
thinking about local content creation and e-commerce as a holistic value chain. This
permits impoverished households to participate as creators of knowledge and generators
of value, not just as consumers, which is currently the case with the Stavrou et al.
(2001) approach.

Local Content Creation in South Africa

Tackling local content and e-commerce presents a challenge for several reasons.
First, significant international attention and debate have focused on regulatory issues
surrounding content, particularly phenomena of child pornography, hate and violent
speech, copyright, ISPs’ liability and advertising. Those issues are not within the
scope of this paper. Second, there is difference in opinion of what is meant by content.
For clarification, this paper recognises content as everything accessible on an electronic
platform (such as a web site) and presented in a range of ways using graphics, text or

183
audio. There are many examples of local content (e.g. learning and teaching resources
or current affairs) but this paper restricts its focus to content from historically
marginalised, cultural and indigenous knowledge systems (IKS) unique to South Africa.
Across the South African telecommunications sectors, a range of departmental
policies and regulations addresses local content creation, but all share commitment to
nation building, unique identity promotion and protection, industrial growth and job
creation (Table 9.5). One of the major players is the regulator, the Independent
Communications Authority of South Africa (ICASA). It offers two main reasons for
regulating local content in South Africa — first, the need to develop, protect and
promote a national and provincial identity, culture and character; and second, the
need to create vibrant, dynamic, creative and economically productive local industries.
The regulator specifies that content needs to “reflect and engage with the life experiences,
cultures, languages, aspirations and artistic expressions that are distinctly South
African”. Local content quotas are set to create incentives for music, television and
broadcasting industries, develop talent, invest in high quality technology, improve the
quality and variety of South African music and television programming, and redress
historical imbalances in the cultural and broadcast industries.

Table 9.5. Governmental and Regulatory Bodies Involved in Local Content Creation
Entity Domains

ICASA TV and radio broadcasting

Language representivity
Quotas for local content
DACST Cultural Industry Growth Strategy
Film, culture and television industries
National Film and Video Foundation
Cultural Heritage Resource Management
DoC Broadcasting and telecommunications
E-Commerce Green Paper (2000)
Support development of local content in broadcasting sector

The Department of Arts, Culture, Science and Technology (DACST), another

player, has been concerned with local content in several areas and several ways. The
Cultural Industries Growth Strategy and the National Film and Video Foundation are
DACST initiatives that focus on local industry growth and stimulating export of South
African film, video and other content as well as promotion of South African talent
and production resources. A third player is the Department of Communications (DoC),
which is concerned with local content creation within its e-commerce and broadcasting
policy processes. The DoC’s White Paper on Broadcasting Policy (1998) includes
regulations on appropriate scheduling of local content, enhanced production quality
for exportation and the establishment of the South African Broadcast Production Agency
to support local content development.

184
 Local content creation across other technological platforms such as the internet
has not received much attention. The only reference to e-content exists in the 1999
amendment to the Film and Publications Act (Act 34 of 1999) and there is no South
African case law on internet content regulation. Some have criticised the apparent
lack of clarity regarding local content and e-commerce and have debated at length
which agencies have jurisdiction to regulate content on the internet. In its discussion
paper on local content creation, ICASA recognises that convergence has an impact on
local content and states that “new communications and information technologies, as
well as the growing convergence between media, entertainment and
telecommunications, pose both a threat to indigenous cultures and an opportunity to
extend the creation and distribution of cultural products and services. The trend
internationally is towards the worldwide homogenisation or standardisation of culture
resulting from the increasing dominance of US cultural content… (L)ocal content
regulation will be severely tested by the emergence of new distribution platforms…”
(http://www.iba.org.za/local_content.doc)

E-Commerce in South Africa

At the time of writing this paper, the South African government is engaged in
compiling a White Paper on e-commerce. Some issues included in the green paper are
development of and access to ICTs, taxation, security and privacy, convergence,
protection of intellectual property, content development and regulation, electronic
payment systems and standards and interoperability. A lot of energy has been directed
at writing into forthcoming legislation what the role of government should be. Different
views obviously exist. Some demand a light touch from government, others want
self-regulation by industry and yet others expect clear policy and regulatory directives
on job creation, alleviation of poverty, black economic empowerment and redressing
inequities. The South African approach to e-commerce shares some similarities with
international approaches and debates, but there are some differences that impact on
the local content creation and e-commerce value chain.

Intellectual Property and Indigenous Knowledge Systems

An environmental difference as well as an enabler relates to Intellectual Property

(IP) and IKS. While the World Intellectual Property Organisation (WIPO) mediates
and arbitrates IP infringements, South African IP faces additional risks from ICTs
because of:
— the legacy of institutionalised destruction of IKS;
— the dominance of Western paradigms in creation and management of IP and
associated value chains;

185
— the absence of adequate legislation relating to the protection of indigenous South
African IP; and
— a poor or limited presence in South Africa of adequate capacity, instruments and
mechanisms to monitor and protect IP rights.
There remains a concern that amendments to South African IP law (the South
African Intellectual Property Law, Act 38 of 1997) do not address adequately all of
the main challenges presented by e-commerce to local content. South Africa participated
in the WIPO IKS discussions and various institutions such as research organisations
now involved in e-commerce development and cultural preservation signalled that
IKS in South Africa had been actively destroyed by apartheid. Some difficulties continue
to exist with respect to IKS and patent protection. For example, foodways, which
include recipes for producing hemp-based health products that were developed
collectively and dynamically over several generations, are not currently protected by
patent law because that knowledge is considered to be in the public domain.

Technologies and Applications in E-Commerce

In the current South African policy process, e-commerce is defined as “the use
of electronic networks to exchange information, products, services and payments for
commercial and communication purposes between individuals (consumers) and
businesses, between businesses themselves, between individuals themselves, within
government or between the public and government and, last, between business and
government” (Department of Communications, 2001). This is a broad definition and
implies a wider range of interactions and players than solely commercial ones.
The technologies involved in facilitating these interactions need to be flagged
because of concerns with affordability and access for impoverished South Africans.
The internet is a major network over which e-commerce interactions occur, but mobile
networks may be more affordable or accessible in the local context. Therefore, adopting
a technology-neutral perspective is necessary. This is especially pertinent in South
Africa where, for example, m-commerce using cellular phones is potentially more
affordable and accessible than e-commerce via personal computers, plain old telephone
systems (POTS) or digital lines.
One way to think about the different dimensions at play across the local content
creation and e-commerce value chain is to refer visually to the application of technology
in context (ATC) model (Eksteen, 2000; Jackson and Eksteen, 2000) (Figure 9.1).
The ATC model has three dimensions, which allow innovation, risk or investment to
be plotted with respect to each dimension. Most developed country e-commerce focuses
on activities in the application-technology plane and on issues such as the hunt for the
next “killer app” or automation of processes. In South Africa, an attempt is made to
focus on the other two planes, namely the context-technology plane and the context-
application plane. Some of the issues that would be considered in the context-technology
plane include combating technological illiteracy or building public understanding of
science and technology.

186
 Figure 9.1. The Application of the Technology in Context Model

Context

Application Technology

Notes:
The technology axis examines issues such as the technologies involved, the degree of
modification or customisation required and so on. For example: Does the technology
exist and is it affordable?
The application axis examines issues such as the underlying business models
involved, the supporting process technologies required and the market opportunities.
For example: Will the application form the core of a small enterprise?
The context axis examines issues such as the needs of the potential users and the
cultural, ethical and social implications of the technologies or application. For
example: Will the technology and application violate a person's right to privacy?

The ATC model is useful for describing visually the different channels,
infrastructures, drivers and enablers that run along the length of the value chain of
local content creation and e-commerce. As the next section of the paper highlights,
approaching these value chains holistically requires attention to context, application
and technology dimensions, and a wide range of issues within the context-technology
and context-application planes.

Local Content Creation and E-Commerce Pilot Research and Development

Projects

South Africa has several research projects and initiatives addressing local content
creation and e-commerce. The selection of projects for mention here is based on their
objectives and the need to illustrate the ATC dimensions of the value chain. Before
describing the pilot projects, one should recognise that important learning regarding
value chains is derived from the open-source movement, and was used in an attempt
to enhance the local content creation and e-commerce value chains underlying the
various projects described in this paper. These lessons were particularly valuable when
local content was sourced from IKS. Although the paper does not explore IKS and
open source in detail, it uses the open-source example of the Linux operating system

187
to illustrate how the success of the open-source model enhanced understanding of the
value chains related to local content creation. IKS and open source have at least three
characteristics in common:
— distributed “ownership” by communities and developers;
— distributed contributions to the whole; and
— creation driven by need, not profit.
Linux is a PC-based operating system produced through a distributed software
development effort. The effort involved more than 3 000 developers and countless
other contributors located in about 90 countries. It is difficult to provide an accurate
figure for the number of programmers who contributed to the Linux operating system
and associated utility programs: published estimates range from several hundred to
more than 40 000 (Shankland, 1998; Raymond, 1999). In its first few years of
development (1991-95), over 15 000 people submitted code or comments to the three
main Linux-related newsgroups and mailing lists. This trend continues today at an
ever-increasing pace.
The simplified value chain (Figure 9.2) begins with the development and creation
of the Linux code. The drivers and value in this activity are typically personal (i.e. non-
financial, like peer recognition) or the result of a specific need for functionality. This
happens in a distributed fashion, and, typically, the Linux community, and not the
individual developer, owns the results. The consumption of the Linux operating system
ranges from the developer community itself to commercial institutions. The Linux
community often takes care of distribution and packaging and uses open systems such
as File Transfer Protocol (FTP) or the World Wide Web.

Figure 9.2. Simplified Value Chain of Linux

Creation, Distribution and Consumption

Increasing Value

$
Needs and Personal Drivers

Red Hat

Linux Market
Development
& Creation Suse Linux

VA Linux

188
 Further, several commercially orientated opportunities that were identified and
then exploited were recognised as located on the border between the developer
community and the broader market. Examples of this include the packaging,
enhancement, distribution and branding of Linux by companies such as VA Linux,
Red Hat and Suse. Although the content included in the “commercial” channels created
by these groups is open source and hence free, it creates value through services and
packaging, enhancement, distribution and branding. This value is then translated into
financial terms.
It is possible to regard these efforts as examples of different channels within the
open-source value chain. The following lessons are relevant:
— Local content creation (like IKS) does not need to happen only to satisfy a
market need. As with Linux, the creation drivers are often not financial or
market orientated.
— Various channels, by adding value, serve to translate local content creation
activities into value measurable in financial terms.
— Not all the channels in the value chain need to be commercially oriented.
— The consumption and creation communities overlap in many instances. With
Linux, the creation community was and still is part of the consumption community.
Exploitation and creation depend on each other. In the Linux case, the balance
between creation and exploitation favours the former. Without sustainable creation
activities, exploitation will fail, even though the creation activities are independent of
the exploitation efforts. This illustrates how different drivers exist at different parts
of the value chain. For ensuring the sustainability of local content value chains it
means that a similar focus is needed to ensure that sustainable and appropriate creation
is available.

Community-Inclusive Value Chains

The Poverty Alleviation through Processing Technologies and Job Creation

Initiative is a multi-million Rand state-funded initiative intended to support rural
people in realising their strengths and potentials through implementation of projects
that result in small and micro enterprises, job creation and poverty relief. The importance
of this initiative, for this paper, is that IKS content and development of products
within an agrarian context can form the basis of a sustainable source of income for
impoverished communities and allow for participation of the poorest of the poor in
South Africa’s e-commerce development. The initiative focuses specifically on value
addition to available local resources, mainly from the agricultural sector but including
IKS. Several projects support the emergence of sustainable value chains for resources
such as hemp, wild silk, leather and indigenous foods. Community participation is
essential in the initiative and community ownership is consolidated in, among other

189
things, establishment of legal and governance entities such as community trusts and
community-owned businesses. This approach, an alternative to supporting individual
entrepreneurs, maximises job creation across the communities.
An activity within this initiative was the design of an architecture for a virtual
trade house which, in itself may not be difficult, but was challenged by the fact that
poverty alleviation and job creation are the overarching objectives of the initiative.
The participatory nature of the initiative was crucial, hence design of a virtual trade
house, for example, had to be consultative and dialogic with the community
stakeholders. Processes and functionality of the virtual trade have to be understood by
all community players — an ongoing challenge because of the low levels of public
understanding of science, technology and e-commerce business models.
In its design, the trade house is not conceptualised as a single entity or as merely
an information technology system. Rather, it is an overall architecture that integrates
systems (e.g. an e-commerce system), activities (e.g. retailing), governance structures
(the board of trustees), legal entities (such as community trusts) and role players
(e.g. creators). Generic and quite typical functionalities are designed into the virtual
trade house, such as distribution, payment, optimisation, centralisation, auditing and
customer account management. A number of unexpected challenges are identified in
the design of the architecture (Table 9.6).
With poverty alleviation and job creation as fundamental objectives, the virtual
aspects of a trade house are not developed using only mainstream e-commerce and
web-enabled commercial marketing approaches. Approaches based on assumptions
and principles of corporate capital accumulation run counter to the principles of this
initiative. Caution has to be applied when commissioning technological implementation
because “off the shelf” products and “mainstream process technologies” may reinforce
accelerated corporatisation.
The livelihoods approach to poverty alleviation underlines that participation is
essential and stresses the need to pay attention to issues such as “acceptance literacy”,
legitimacy and buy-in of both the specific projects and the overall initiative. The
sustainability of the value chain is put at risk unless there is attention to ongoing
training and awareness, as well as to governance and communication, which ensure
that breaks or blockages in the value chain do not result and that the communities
involved are not alienated.

190
 Table 9.6. Problems Met in Designing a Trade House Architecture

Branding There is a need to decide on branding across several entities ranging from, for example “[brand name] hemp” or
“[community name] product” or “[community trust name] products” or “Collective virtual trade house [name]”. The
brand management process is complicated and decisions still need to be made about the level at which branding is
required. The decision, however, has an impact on what income from royalty payments, for example, can be generated
by participating communities.
Quality control Quality control impacts more on certain products than on others, particularly those that use IK. For example, hemp
products used for health care may need to be quality controlled for toxicity, and an “external” quality control process
and associated disclaimers need to be visible on products. The South African Bureau of Standards (SABS) approval or
a “fair trade” approval may add to the value of the product and perceptions of brand purity. This process, however,
presents a risk that IK involving traditional healing processes is transformed, which runs counter to preserving that
knowledge.
Financing Refers to consequences associated with success, namely that as a legal entity the trade house may have enhanced
access to external financing options or may itself evolve into a financial service provider. This raises the possibility of
development of the “next generation” community bank, providing financing to impoverished households or SMMEs.
191

Hence, it transforms the trade house into a legitimate micro-lending entity, which could be a route for further
development.
Depersonalisation This is a risk associated with success because as the community collective becomes more successful, it may reach a
transition point from which its personalised and community-flavoured services and products are lost as it transforms
into a quasi or real corporate, even without any changes in its legal composition or governance.
Knock outs A risk that some products or product ranges such as hemp clothing and hemp-based health care products become so
successful that they knock out the smaller and less popular products such as wild silk linens. This would run counter to
the principle that the overall collectively owned trade house enhances value for all.
 Digitising Indigenous Cultural Content

A second project is the Culture Preservation Project, also a state-funded multi-

million Rand entity. Among the several publications available that outline the project’s
objectives and its innovations in methodologies and technologies, Jackson and Eksteen
(2000) discuss IKS and the impact that globalisation and ICTs have on its management.
IKS is understood dialectically as a) a recognition that indigenous people are inventors
and custodians of technology and knowledge, and b) having multiple manifestations,
many of which are directly interlinked in communal cultural practices, traditions
and symbols.
Because IKS is a rich source of local e-content, the impact of digitisation on
that e-content needs understanding. Commonly used technologies include digitisation
technologies, 3D modelling tools, encryption algorithms, digital libraries and so on.
From a purely technological perspective, once digitisation and technological
communication functionality are in place, little or no technological modification is
required to set up a B2C e-commerce service. The actual impacts of ICTs on IKS are
not always positive. Current ICTs are applied largely to support global commercial
transactions and as a result, knowledge — indigenous or otherwise — is a commodity,
potentially controlled and sold without attention to cultural preservation.
In IKS, a simple technological intervention such as digitally watermarking a
digitised cultural artefact can have unintended cultural consequences. A digital
watermark is a pattern of binary digits inserted into the artefact that provides
information about copyright. It is not visible and must be robust enough to continue
to exist if the digital artefact is changed in some way. Digital watermarking cannot
guarantee the authenticity of a cultural artefact in cultural terms (i.e. that the e-content
is an authentic representation of a culture). It provides a trace of origin of ownership.
It contributes towards the commodification of IKS-based e-content and its management.
The addition of a watermark can be perceived to provide authenticity. In those cases,
the perception is very different from real authenticity. It is not inconceivable that
perceptions can be manipulated or misrepresented, particularly if the e-content supports
purely short-term commercial interests.
IKS e-content can be used for various purposes, such as education, cultural
preservation, public awareness and virtual cultural tourism. The principles of its digital
protection and preservation rest on enhanced restitution and human worth, and permit
current and future communities to have continued and authentic access to the symbols,
values and traditions which shape and define them as unique. The act of preserving
digitally cultural artefacts is not a once-off activity. It is ongoing, requiring investment
and commitment and it is at this point that concerns with “sustainability” often arise.
One view suggests allocation of finite state funding based on the reasoning that the
state must “manage” IKS on behalf of the people. The risk is that state funding can
change or disappear, particularly when priorities, policies or social conditions change.
A second view opts for “soft commercialisation”, where not-for-profit income for
particular services is rechannelled into purely preservation activities. Two additional

192
views suggest doing nothing at all or pursuing “hard commercialisation” — wholesale
commodification of knowledge based on the rules of the game defined by the global
market place.
With respect to creating channels for use of local e-content in the global market
place, some regard cultural tourism as a soft commercial application that balances
tensions between preserving indigenous cultures and economic reconstruction and
development. Cultural tourism, as defined by the World Trade Organisation, is a
process whereby an individual is motivated to travel to a different country and culture
for aesthetic, intellectual or spiritual development. Yet it also is a business transaction
and a marketing exercise, facilitated by ICTs. The act of preservation of the indigenous
culture for use as local e-content is a necessary condition, but there is a risk that
investors may want higher financial returns on the investment made in preservation.
Virtual cultural tourism is a growing market, but digitised IKS for commercial
e-consumption could present cultures in a trivialised or paternalistic way. In designing
a compelling cultural virtual website, one challenge is to avoid such trivialisation.
Retaining authenticity without making the experience compelling risks losing
consumers who want more excitement. Tension rules between market need and
preservation.
Sourcing IKS for e-content risks reinforcing the historical marginalisation of
indigenous people and their knowledge. Digitisation of indigenous content demands
sensitivity to the historical and human systems within which it exists. It also requires
high technical skills. Given that the artefact is neither neutral nor ahistorical, digitisation
cannot occur without understanding the cultural and curatorial aspects at play.
Local content regulation and management are complicated because digitisation
creates a new artefact and not merely a reproduction of the original. As a result,
different rules and information management processes (subject to different perceptions)
come into play. A way of ensuring that the creators of knowledge accrue benefit,
either financially or by acknowledgement, has been formulated and the establishment
of a Universal Digital Original (UDO) master index suggested. The UDO concept
involves a high-quality master digital copy or representation of an artefact, concept,
event or natural phenomenon. The UDO would meet certain international standards
for digitisation and be accompanied by an authenticity certificate. UDOs could be
registered and managed, possibly by or on behalf of the state, and would list licence
agreements as well as commercial and non-commercial rights as defined by use
regulations. This would make it possible to address some of the issues of compensation
or recognition for communities involved in IKS-based e-content supply, so that they
are able to support preservation of IKS.
The view offered here has highlighted some technological, context and application
considerations in the local content creation and e-commerce value chain, especially
when one of the channels is cultural tourism. Some of the solutions developed may
appear to observers from developed countries as too cumbersome or to have too much
governmental involvement, but from a developing country perspective, as in South
Africa, a wider range of considerations than only market demand has to be considered.

193
A Possible Future Instantiation of Value Chains

Impoverished households have assets and abilities but simultaneously face a

diversity of risks affecting their options and opportunities for securing sustainable incomes
(Stavrou et al., 2001). Most, particularly rural ones, are headed by women, many of
whom are grandmothers performing childcare work without remuneration and reliant
on income from social-support grants (pensions, disability and child support grants) or
ad hoc employment by some or all members of the household, including children.
The Storytellers’ Network Initiative, the third project for discussion here, is
conceptualised within this context. It seeks to source stories and knowledge embedded
in the daily lives, activities and memories of impoverished community members,
particularly (but not exclusively) older women, in order to provide opportunities for
commercial and non-commercial use of that content. South Africa has a rich history
of storytelling, poetry, music and creative expression. The initiative assumes that supplying
local content for development into e-content is one way that impoverished households
and communities can simultaneously participate in South Africa’s e-commerce
development and secure additional and sustainable income sources for themselves.
Trained community members source the local content in the community. Cultural
authenticity is established through rigorous yet creative mechanisms. Digitisation of
the content follows guidelines specified according to international approaches that
seek to preserve cultural content with technical and cultural care. One of the channels
available for creating income-generating opportunities uses the content to develop
IKS-based early childhood development programmes. Early childhood development
(ECD) is an educational and developmental imperative. South African psychological
and educational literature recognises that the notions of childhood and development
are not compatible with traditional, unquestioned assumptions that development is
“coherent”. The ECD channel seeks to source knowledge about child development as
well as content and tools, such as stories, games, indigenous toys and play, across a
range of cultural systems indigenous to South Africa. In combining educational,
ethnographic and anthropological research and community participation in sourcing
indigenous cultural childrearing practices, rich ECD learning content and programmes
emerge. The community itself can benefit both from the sourcing function it plays
and from income created for women who establish ECD centres in their communities.
Future plans exist for supplying content to South Africa’s publishing, broadcasting
and tourism industries as well as providing exportable ECD content in an electronic
format (e.g. on CD and DVD) from which royalty fees will be available to the community.
A second available channel sources content as the basis of printed and electronic
material, such as postcards, poetry anthologies, story anthologies, historical descriptions
and product-packaging descriptions of other products from the region. The arts and
crafts channel co-packages this content with hand-made arts and crafts created in the
region and allows households or groups of them to provide products for cultural
stores, both physical and virtual. An option provides these as corporate gifts, which
South African organisations can purchase and distribute to international and local

194
clients. The Storytellers’ Network Initiative also faces “acceptance literacy” problems
among community participants and the risk of “knock outs”, as well as issues about
the impact of digitisation of indigenous content similar to those faced by the Culture
Preservation Project.
As already noted, the use of cellular or mobile telephones is increasing in South
Africa. The behavioural patterns of using that infrastructure are very interesting.
South Africans, like their counterparts in many other countries, use SMS functionality
on their cellular telephones in very significant proportions. Its rapid increase signals a
trend that people want to and do use their cellular phones primarily to connect with
other people. This suggests that content may not be as dominant in the future as it is
currently. Focus will go instead on providing infrastructure, channels and services
that connect people who are “consumers” directly with those who are creators. At
present, content acts in an intermediary way between consumers and creators but it is
likely that social networks rather than distribution and intermediary networks may
become dominant. These value chains may thus become predominantly social ones,
with a challenge to the Storytellers’ Network Initiative to ensure that the value chain
can evolve towards a social one. Different drivers, channels, infrastructure and enablers
will come into play across the length of the value chain. As with the open-source
example, several commercially orientated opportunities may appear either on the border
between consumers and creators or in totally unexpected locations. This development
will not be without risk. South Africa could take the lead in exploring future instantiations
of local content and e-commerce value chains. Even if environmental factors such as
poverty or inequitable access to ICTs prevail (but are much reduced, one may hope)
future concepts of value chains could become shaped more by humanism and people-
driven communication than currently.

Conclusion

This paper has attempted to present thinking about local content creation and e-
commerce in terms of holistic value chains that span both virtual and physical
dimensions and stretch across a wider range of activities than simple consumption. It
suggests that a number of different channels, infrastructures, drivers and enablers run
along the length of the chains and differ considerably between different points on them.
Several local content creation and e-commerce initiatives are underway in South
Africa. The paper discusses three of them. Drawing on local content sourced from
IKS and agrarian, rural and predominantly impoverished settings, it showed how the
poorest of the poor can indeed participate as more than merely consumers. It referred to
a possible future instantiation of value chains and the need for urgency — in developing
countries — to accelerate research and pilots on alternative and innovative local content
and e-commerce value chains, premised on principles not exclusively commercial.

195
 Note

1. Subsequent to the publication of this paper, the White Paper process has resulted in
the formulation of the current Draft Electronic Communication and Transactions
Bill (http://www.gov.za/gazette/bills/2002/23195.pdf).

196
 Bibliography

AFRICAN INTERNET CONNECTIVITY, http://www3.sn.apc.org/africa/, last accessed 6 April 2001.

DEPARTMENT OF COMMUNICATIONS http://www.doc.pwv.gov.za last accessed 6 April 2001.
EKSTEEN, J (2000), The Application of Technology in Context Model, unpublished paper,
CSIR, Pretoria.
H ALL, M. (2001), Africa Connected, First Monday, http://www.firstmonday.dk/issues/
issue3_11/hall/index.html, last accessed 19 April 2001.
INDEPENDENT COMMUNICATIONS AUTHORITY OF SOUTH AFRICA (2001), Discussion Paper on the
Review of Local Content Creation, http://www.iba.org.za/local_content.doc, last
accessed 6 April 2001.
JACKSON, C.-A. and J. EKSTEEN (2000), Device-ive Artefacts, paper presented at the South
Africa — China Indigenous Knowledge Systems Summit, Beijing, China, October.
NETWORK WIZARDS (2000), cited in HALL (2001).
RAYMOND, E.S. (1999), The Cathedral and the Bazaar: Musings on Linux and Open Source
by an Accidental Revolutionary, O’Reilly & Associates, Sebastopol, CA.
SHANKLAND, S. (1998), Linux Shipments up 212 per cent, available at CNET News.com
http://news.cnet.com/category/0-1003-200-336510.html, last accessed
19 April 2001
STATISTICS SOUTH AFRICA (1997), http://www.statsa.gov.za, last accessed 6 April 2001
STAVROU, A, J. MAY and P. BENJAMIN (2001), “E-Commerce And Poverty Alleviation In South
Africa: An input paper to the Government Green Paper on E-Commerce”, University
of Natal, Durban.
UNECA (2000), cited in HALL (2001).

197
LIST OF PARTICIPANTS
AND
CONFERENCE AGENDA

199
 List of Participants

Salahuddin Aminuzzaman Professor, University of Dhaka, Bangladesh

Pier Giorgio Ardeni President, SDIC-University of Bologna, Italy
Annaflavia Bianchi Fellow, Telecom Italia, Italy
Jorge Braga de Macedo President, OECD Development Centre, France
Kyle Eischen Director of Programs and Development, CGIRS-UCSC,
United States of America
Johan Eksteen Research Fellow, CERNA-École des Mines, France
Patrizia Fariselli Consultant, Nomisma, Italy
Gary Gereffi Professor, Duke University, United States of America
Andrea Goldstein Senior Economist, OECD Development Centre, France
Paolo Guerrieri Professor, University of Rome, Italy
David Hallam Division Chief, Food and Agriculture Organisation, Rome
Brian Hammond Division Chief, OECD Development Co-operation
Directorate, France
John Humphrey Fellow, Institute for Development Studies, United Kingdom
Carey-Ann Jackson Consultant, South Africa
Gilles Leblanc Research Fellow, CERNA-École des Mines, France
Aaditya Mattoo Senior Economist, World Bank, United States of America
Sagren Moodley Researcher, CSDS-University of Natal, South Africa
Lynn Mytelka Director, UNU-INTECH, Netherlands
Catherine Nyaki Adeya Research Fellow, UNU-INTECH, Netherlands
David O’Connor Senior Economist, OECD Development Centre, France
Danilo Piaggesi Division Chief, IADB, United States of America
Fabiola Riccardini Economist, ISTAT, Italy
Morten Scholer Senior Adviser, International Trade Centre, Switzerland
Susanne Teltscher Economic Affairs Officer, UNCTAD, Switzerland
Paul Timmers Head of Sector, DG INFSO, European Commission,
Belgium
Mark Tomlinson Research Fellow, University of Manchester, United
Kingdom

201
 4 May 2001

9:30 WELCOME ADDRESS

Pier Ugo Calzolari, Rector of the University of Bologna

10:00 INTRODUCTION TO THE WORKSHOP

Pier Giorgio Ardeni (Director, SDIC)
Jorge Braga de Macedo (President, OECD Development Centre)

Session 1: General Issues

10:15 E-COMMERCE FOR DEVELOPMENT

Patrizia Fariselli (SDIC)
David O’Connor (OECD Development Centre)
Discussant: Lynn Mytelka (UNU-INTECH, Maastricht)

12:00 THE IMPACT OF THE INTERNET ON GLOBAL VALUE CHAINS

Gary Gereffi (Duke University)
Discussant: Annaflavia Bianchi (Telecom Italia, Venice)

Session 2: Learning from Best Practices

14:30 MEASURING E-COMMERCE IN DEVELOPING COUNTRIES

Susanne Teltschler (UNCTAD, Geneva)
Discussant: Fabiola Riccardini (ISTAT, Rome)

15:15 ACCESS TO TELEPHONY AND THE INTERNET: THE BANGLADESHI EXPERIENCE

Salahuddin Aminuzzaman (University of Dhaka) & Ishtiaq Jamil
(University of Bergen)
Discussant: Mark Tomlinson (University of Manchester)

16:15 INTERNATIONAL COOPERATION

Danilo Piaggesi (Inter-American Development Bank, Washington)
Paul Timmers (European Commission, Brussels)
Discussant: Brian Hammond (OECD Development Co-operation
Directorate)

202
 5 May 2001

9:00 TRADE POLICIES FOR ELECTRONIC COMMERCE

Aaditya Mattoo (World Bank, Washington)
Discussant: Paolo Guerrieri (University of Rome)

Session 3: Case Studies: Part I

9:45 MANUFACTURING
GLOBAL SUPPLY CHAINS IN THE CAR INDUSTRY
Andrea Goldstein (OECD Development Centre,)
Sagren Moodley (University of Natal, Durban)
Discussant: John Humphrey (IDS, University of Sussex, Brighton)

10:45 PRIMARY COMMODITIES

THE IMPACT OF INTERNET TRADING IN THE COFFEE INDUSTRY
Morten Scholer (International Trade Centre, Geneva)
Discussant: David Hallam (Food and Agriculture Organisation,
Rome)

Session 4: Case Studies: Part II

11:45 E-COMMERCE AND THE ICT INDUSTRY: IMPACT ON LOCAL DEVELOPMENT IN INDIA
Kyle Eischen (CGIRS, UC Santa Cruz)
Discussant: Gilles Le Blanc (STICERD-LSE & École des Mines,
Paris)

12:30 E-COMMERCE & LOCAL CONTENT CREATION IN SOUTH AFRICA

Carey-Ann Jackson (Consultant, South Africa)
Discussant: Catherine Nyaki Adeya (UNU-INTECH, Maastricht)

13:30 Closing Address

Fabio Roversi Monaco, (President, Fondazione CARISBO, Bologna)

203
OECD PUBLICATIONS, 2, rue André-Pascal, 75775 PARIS CEDEX 16
PRINTED IN FRANCE
(41 2002 09 1 P) ISBN 92-64-09954-9 – No. 52759 2002